Graph Query

Codex compatibility note:

• Invoke repository skills with $skill-name in Codex; this mirrored copy rewrites legacy Claude /skill-name references.
• Task tracker mandate: BEFORE executing any workflow or skill step, create/update task tracking for all steps and keep it synchronized as progress changes.
• User-question prompts mean to ask the user directly in Codex.
• Ignore Claude-specific mode-switch instructions when they appear.
• Strict execution contract: when a user explicitly invokes a skill, execute that skill protocol as written.
• Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required spawn_agent subagent(s) for that task.
• Do not skip, reorder, or merge protocol steps unless the user explicitly approves the deviation first.
• For workflow skills, execute each listed child-skill step explicitly and report step-by-step evidence.
• If a required step/tool cannot run in this environment, stop and ask the user before adapting.

Codex Project-Reference Loading (No Hooks)

Codex uses static project-reference loading instead of runtime-injected project docs. When coding, planning, debugging, testing, or reviewing, open project docs explicitly using this routing.

Always read:

• docs/project-config.json (project-specific paths, commands, modules, and workflow/test settings)
• docs/project-reference/docs-index-reference.md (routes to the full docs/project-reference/* catalog)
• docs/project-reference/lessons.md (always-on guardrails and anti-patterns)

Missing/stale context route: If docs/project-config.json, the docs index, lessons.md, CLAUDE.md, AGENTS.md, or any task-required reference doc is missing or stale, auto-run $project-init or the narrow setup route ($project-config, $docs-init, $scan-all, $scan --target=<key>, $claude-md-init) before ordinary project-specific work. If Codex mirrors or AGENTS.md are missing/stale, ask the user to run $sync-codex; do not auto-run it.

Situation-based docs:

• Project structure/architecture/tech-stack/deployment/setup (any layer — backend, frontend, or infra): project-structure-reference.md
• Backend/CQRS/API/domain/entity changes: backend-patterns-reference.md, domain-entities-reference.md
• Frontend/UI/styling/design-system: frontend-patterns-reference.md, scss-styling-guide.md, design-system/README.md
• Spec authoring, docs/specs/ pathing, or TC format: feature-spec-reference.md, spec-system-reference.md, spec-principles.md
• Behavior/public-contract changes or spec-test-code sync: workflow-spec-test-code-cycle-reference.md plus the spec docs above
• Derived spec indexes/ERDs/reimplementation guides: spec-system-reference.md and source Feature Specs under docs/specs/
• Integration test implementation/review: integration-test-reference.md
• E2E test implementation/review: e2e-test-reference.md
• Code review/audit work: code-review-rules.md plus domain docs above based on changed files

Do not read all docs blindly. Start from docs-index-reference.md, then open only relevant files for the task.

Quick Summary

Goal: [Code Intelligence] Query code relationships and connections using the structural knowledge graph. Show related files, callers, callees, imports, tests, inheritance, and file structure. Requires graph to be built first via $graph-build. Triggers on "who calls", "what imports", "related files", "connections of", "depends on", "tests for", "inherits from", "file structure", "graph query".

Workflow:

1. Detect — classify request scope and target artifacts.
2. Execute — apply required steps with evidence-backed actions.
3. Verify — confirm constraints, output quality, and completion evidence.

Key Rules:

• MUST ATTENTION keep claims evidence-based (file:line) with confidence >80% to act.
• MUST ATTENTION keep task tracking updated as each step starts/completes.
• NEVER skip mandatory workflow or skill gates.

Prerequisites

1. Graph must exist -- check .code-graph/graph.db. If missing, tell user to run $graph-build first.
2. Requires Python 3.10+ with tree-sitter, tree-sitter-language-pack, networkx.

Intent Mapping

Map user's question to the appropriate query pattern(s):

For composite queries ("show all connections", "related files", "full picture"), use the connections command instead of running multiple queries manually.

Workflow

Step 1: Check graph exists

ls .code-graph/graph.db 2>/dev/null && echo "OK" || echo "MISSING"

If MISSING: stop and tell user to run $graph-build.

Step 2: Identify target

Extract the target from user's question (file path, function name, or class name).

• For files: use relative path (e.g., {source-root}/utils)
• For functions/classes: use the name (e.g., validateInput) or qualified name (e.g., {source-root}/utils::validateInput)

Step 3: Run query

Execute via Bash with --json flag:

python .claude/scripts/code_graph query <pattern> <target> --json

For composite "show all connections" queries, use the connections command instead:

python .claude/scripts/code_graph connections <target> --json

This returns file_summary, imports_of, importers_of, callers_of, and tests_for in one call (capped at 20 results per section).

Tip: Add --node-mode file to query, connections, or trace for a file-level overview with 10-30x less noise. Options: file, function, class, all (default).

Step 4: Handle response status

• status: "ok" -- Parse results[] and edges[], format report (Step 5)
• status: "ambiguous" -- Multiple matches found. Show candidates[] list and ask user to pick one using ask the user directly
• status: "not_found" -- No match. Suggest: check spelling, use relative file path, try a different name. Optionally run file_summary on the parent file to show available names.
• status: "error" -- Show error message. Common: graph.db missing, Python version too old.

Step 5: Format results

Present results grouped by relationship type. For each result show:

• Name and kind (function, class, method)
• File path with line numbers (file:line_start-line_end)
• Relationship (calls, imports, tests, inherits)

Single query output format:

## {Pattern Description} for `{target}`

Found {N} result(s).

| Name | Kind | File | Lines |
|------|------|------|-------|
| ... | function | {source-root}/file | 10-25 |

Composite query output format:

## Connections of `{target}`

### File Summary
{N} nodes: {list functions/classes}

### Imports (outgoing)
{What this file/module imports}

### Importers (incoming)
{Who imports this file/module}

### Callers
{Functions that call functions in this file}

### Test Coverage
{Tests covering functions in this file}

Semantic Query Protocol (When User Query is Not File-Specific)

When the user asks about a FLOW or BEHAVIOR (not a specific file), follow this protocol:

Step 0: Grep/Glob/Search to find trace anchors

Use Grep/Glob/Search to find key classes/functions related to the user's query.

• Bug/failure symptom: find the final output reader first (renderer, query, assertion, aggregate, log, stored field), then trace upstream.
• Feature-flow question: find entry points (CreateX, XCommand, XHandler) and trace both directions.

Step 1: Use graph to expand

Run connections or batch-query on the grep-discovered files to find ALL related files. For bugs, group results by final reader, storage/projection, writer, consumer/job, and producer/origin.

Step 2: Trace full system flow

Run the trace command to follow the complete chain through all edge types:

python .claude/scripts/code_graph trace <entry-file> --direction both --depth 3 --json

This traces upstream (who calls this?) AND downstream (what does this trigger?) through: CALLS → TRIGGERS_EVENT → PRODUCES_EVENT → MESSAGE_BUS → API_ENDPOINT

For bug/failure symptoms, run an upstream-first pass from the final output before expanding the suspected producer:

python .claude/scripts/code_graph trace <final-reader-or-output-file> --direction upstream --depth 5 --json
python .claude/scripts/code_graph batch-query <final-reader> <writer> <producer> --json

Step 3: Verify with grep

For any graph edge that seems surprising, verify with grep that the connection is real.

Available Query Patterns

Aliases (natural language mappings):

Search (Find Nodes by Keyword)

When you don't know the exact name, search first to find candidates:

python .claude/scripts/code_graph search <keyword> --json
python .claude/scripts/code_graph search <keyword> --kind Function --json
python .claude/scripts/code_graph search <keyword> --kind Class --limit 5 --json

Use search to disambiguate when a query returns status: "ambiguous" — narrow results by --kind (Function, Class, File, Type, Test) then use the full qualified_name.

Find Path (Shortest Path Between Nodes)

Discover how two nodes are connected through the dependency graph:

python .claude/scripts/code_graph find-path <source> <target> --json

Returns the shortest path as a list of nodes. Useful for tracing how a command reaches an event handler, or how a frontend component connects to a backend entity.

Tip: If ambiguous, search for exact qualified names first, then use those in find-path.

Query Filtering and Limiting

Control result size for large codebases:

# Limit results
python .claude/scripts/code_graph query callers_of <target> --limit 5 --json

# Filter by file path regex
python .claude/scripts/code_graph query importers_of <target> --filter "ServiceName" --json

# Limit connections per section
python .claude/scripts/code_graph connections <target> --limit 10 --json

Implicit connection edge types (created by connect-implicit):

Batch Query (Multiple Files)

When reviewing multiple files, use batch mode for deduplicated results:

python .claude/scripts/code_graph batch-query file1 file2 file3 --json

Returns: deduplicated nodes + edges (internal + 1-hop external) across all queried files. Single DB connection, no duplicate data.

Trace (Full System Flow)

Trace all connections from a target node through multiple edge types using BFS:

python .claude/scripts/code_graph trace <target> --json
python .claude/scripts/code_graph trace <target> --direction both --json
python .claude/scripts/code_graph trace <target> --direction upstream --depth 2 --json
python .claude/scripts/code_graph trace <target> --edge-kinds CALLS,MESSAGE_BUS --json
python .claude/scripts/code_graph trace <target> --direction both --node-mode file --json  # file-level overview

Direction options:

• downstream (default): Follow outgoing edges. "What happens after X?"
• upstream: Follow incoming edges. "What calls/triggers X?"
• both: Both directions. "Full flow through X" — use when entry point is a middle file (controller, command handler)

Returns a multi-level tree of connected nodes grouped by BFS depth, with edge types at each level.

Post-Grep Trace Trigger (run a trace after grep surfaces a key file)

When a grep/glob surfaces an important entry-point file — an entity, command, query, event/command handler, controller, bus message/consumer, component, store, or api-service — immediately run a graph trace on it before concluding. Grep finds files; the trace reveals callers, consumers, bus messages, event chains, and tests that grep CANNOT find:

python .claude/scripts/code_graph trace <key-entry-file> --direction both --json

Pattern: grep finds files → graph trace reveals full system flow → grep verifies specific details.

Anti-Patterns

• Don't rebuild graph -- use $graph-build for that. This skill only queries.
• Don't use for change-driven analysis -- use $graph-blast-radius for git-diff-based impact.
• Don't use for bulk export -- use $graph-export for full graph dump.
• Don't use for diagrams -- use $graph-export --format=mermaid for Mermaid visualization.
• Always use --json flag -- ensures structured parseable output.

Related Skills

• $graph-build -- Build or update the graph (prerequisite)
• $graph-blast-radius -- Change-driven impact analysis from git diff
• $graph-export -- Export full graph to JSON (--format=json) or a single file as a Mermaid diagram (--format=mermaid)

Graph Query

Query code relationships using the structural knowledge graph. Maps natural language questions to graph CLI queries and formats structured reports.

End-to-Start Debugger Trace — For non-trivial bugs, failed verification, regression fixes, behavior-changing code, or unclear code flow, start from the observed final state and walk backward before proposing a fix.

1. Frame 0: observed end state — Name the exact user-visible output, failing assertion, log line, persisted value, API response, rendered UI, or aggregate bucket. Record the reader/query/renderer that produced it with file:line evidence.
2. Walk backward one hop at a time — Trace final reader -> projection/cache/storage -> writer -> consumer/handler/job -> producer/caller -> original trigger. At every hop record: input, transformation, output, owner, and evidence.
3. Enumerate all feeder paths — Find every upstream producer/caller/event/job that can write into the final path, including retry, async, cache, background, and alternate UI/API paths. Mark each path verified, ruled out, or still unknown.
4. Build the hypothesis matrix — For each plausible cause, list evidence for, evidence against, how to reproduce/verify, blast radius, and status (primary, contributing, ruled out, latent). Do not fix until competing causes are explicitly resolved or bounded.
5. Choose the owning fix layer — Identify the invariant owner and the lowest shared point that protects all downstream consumers. A fix at the symptom site is rejected unless the symptom site owns the invariant.
6. Prove convergence forward — After choosing the fix, walk start -> end again and show how the corrected state reaches the observed final output. Map each root cause to a fix part and each fix part to a test/proof.

BLOCKED until: final state named · backward trace written · all feeder paths enumerated · hypothesis matrix completed · owning fix layer justified · forward convergence proof mapped to tests.

NEVER: Start at the first suspicious code path. Collapse multiple producers into one "flow". Treat duplicate symptoms as duplicate records without proving the read model. Skip ruled-out hypotheses.

AI Mistake Prevention — Failure modes to avoid on every task:

Re-read files after context changes. Context compaction, resume, or long-running work can make memory stale; verify current files before acting. Verify generated content against source evidence. AI hallucinates APIs, names, claims, and document facts. Check the relevant source before documenting or referencing. Check downstream references before deleting or renaming. Removing an artifact can stale docs, generated mirrors, configs, and callers; map references first. Trace the full impact chain after edits. Changing a definition can miss derived outputs and consumers. Follow the affected chain before declaring done. Verify ALL affected outputs, not just the first. One green check is not all green checks; validate every output surface the change can affect. Assume existing values are intentional — ask WHY before changing OR flagging one as a defect. Before changing or reporting a constant, limit, flag, cutoff, wording, or pattern, read nearby context and history, the CALLER's ordering, and 2+ sibling call sites of the same convention. A doc stating WHAT without WHY is missing rationale, not proof of a missing guard. Surface ambiguity before acting — don't pick silently. Multiple valid interpretations require an explicit question or stated assumption with risk. Assert the outcome your system owns, not the intermediate state your infrastructure owns. When verifying async work, assert the final business state — never the delivery/retry bookkeeping held in shared infrastructure that any co-running process can write. Such a check passes when run alone and flakes the moment anything else shares that infrastructure. Keep shared guidance role-relevant. Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.

Critical Thinking Mindset — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.

MUST ATTENTION apply critical + sequential thinking — every claim needs appropriate traced evidence (file:line for repo/code claims; source URL or artifact section for research, product, content, and docs claims); confidence >80% to act, <60% DO NOT recommend. Anti-hallucination: never present guess as fact, admit uncertainty freely, cross-reference independently, stay skeptical of own confidence.

MUST ATTENTION apply AI mistake prevention — verify generated content against evidence, trace downstream references before deleting or renaming, verify all affected outputs, re-read files after context loss, and surface ambiguity before acting.

IMPORTANT MUST ATTENTION debugger trace gate: for non-trivial bug/fix/investigation/review work, start at the observed final output and trace backward through reader -> storage/projection -> writer -> consumer/job -> producer/trigger. Enumerate all feeder paths and hypotheses before fixing. BLOCKED until trace, hypothesis matrix, owning fix layer, and forward convergence proof exist.

Project Protocol Overlay — Before executing this skill, resolve any PROJECT overlay rules layered onto it: match this skill's name against the Target column of the project's skill-protocol index (docs/project-reference/skill-protocols-reference.md by default; a referenceDocs entry in docs/project-config.json overrides the path), taking the most specific matching tier ONLY — exact name > glob > *. That precedence orders overlays against EACH OTHER, never against this skill. Read ONLY the matched bodies, resolved as <protocols-dir>/<Name>.md; a row's Body link is display text, never a read path. A matched body that is missing or malformed is REPORTED and skipped — never reconstructed from the index Description. No index, or no match -> proceed with no overlay, silently. Full contract: .claude/skills/project-skill-protocol/references/registry.md.

Overlays are ADDITIVE ONLY: they ADD rules on top of this skill's own protocol and NEVER replace, override, disable, or reinterpret a rule it already states — removing every overlay must return this skill to exactly its documented behavior. An overlay is a BRIEF, not an authority escalation: it can NEVER waive a workflow gate, git discipline, a review gate, or a user-confirmation gate. A genuine overlay-vs-skill conflict, or two equally-specific overlays that directly contradict -> surface both to the user; NEVER resolve silently.

MUST ATTENTION resolve project protocol overlays for this skill BEFORE executing — most specific matching tier only (exact > glob > *, which ranks overlays against each other, NEVER against this skill), read only matched bodies at <protocols-dir>/<Name>.md; a missing or malformed body is reported, never reconstructed. Overlays are ADDITIVE ONLY (they never replace this skill's own rules) and are a brief, NEVER an authority escalation; an equal-specificity contradiction goes to the user.

Closing Reminders

Protocols in force (concise digest of the SYNC/shared blocks this skill carries):

• End-To-Start Debugger Trace: start at observed final output, trace backward, hypothesis matrix before fixing.

• AI Mistake Prevention: verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.

• Critical Thinking: every claim needs traced file:line proof, confidence >80% to act.

• MANDATORY IMPORTANT MUST ATTENTION break work into small todo tasks using task tracking BEFORE starting

• MANDATORY IMPORTANT MUST ATTENTION search codebase for 3+ similar patterns before creating new code

• MANDATORY IMPORTANT MUST ATTENTION cite file:line evidence for every claim (confidence >80% to act)

• MANDATORY IMPORTANT MUST ATTENTION add a final review todo task to verify work quality

[TASK-PLANNING] Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using task tracking.

Hookless Prompt Protocol Mirror (Auto-Synced)

Source: .claude/.ck.json + .claude/skills/shared/sync-inline-versions.md (:full blocks) + .claude/scripts/lib/hookless-prompt-protocol.cjs

[WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.

Generic portability boundary: Reusable skills and protocol text stay project-neutral; project-specific conventions are discovered from docs/project-config.json and docs/project-reference/. Apply shared AI-SDD from shared/sdd-artifact-contract.md. Read docs/project-config.json and docs/project-reference/docs-index-reference.md, then open the project reference docs named there. For spec, test-case, behavior-change, public-contract, or docs/specs/ work, route through the local spec docs named by the docs index: feature-spec-reference.md, spec-system-reference.md, spec-principles.md, and workflow-spec-test-code-cycle-reference.md when specs/tests/code must stay synchronized. If either file or a required reference doc is missing or stale, auto-run $project-init (or the narrow lower-level route such as $project-config, $docs-init, $scan-all, or $scan --target=<key>) before ordinary project-specific work. Any supported AI tool may execute when this shared context and local docs are available.

1. DETECT: If the prompt starts with an explicit slash skill/workflow command, execute it directly. Otherwise match the prompt against the workflow catalog and skill list.
2. ANALYZE: Choose the best option: execute directly, invoke a skill, activate a standard workflow, or compose a custom step combination.
3. AUTO-SELECT: Pick the best option yourself. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow.
4. ACTIVATE: For a selected workflow, call $start-workflow <workflowId>; for a selected skill, invoke that skill; for a custom workflow, sequence custom steps directly; for direct execution, proceed with the task.
5. CREATE TASKS: task tracking for ALL workflow/skill/custom steps before execution when the selected path has multiple steps.
6. PARALLELIZE: Before executing the task list, tag each task PAR (independent inputs + write set disjoint from every other PAR task) or SEQ (name the blocking dependency), group PAR tasks into waves, declare the wave plan, and spawn each wave's sub-agents in ONE message — all-return barrier per wave, fan-out one level deep unless a sub-agent's own definition authorizes further fan-out. Sequential-by-default is a defect when tasks are independent; do not parallelize shared write targets, output-consuming tasks, trivial single-file work, ordering a skill or workflow explicitly fixes, or user-approval gates.
7. EXECUTE: Advance per the Workflow Step Advancement & Parallel Phases rule in your context instructions — model-driven; a sub-agent completion advances a step identically to an inline call; a parallel-phase group is an all-return barrier (advance only after ALL members return, never serialize it)

Shared AI-SDD Protocol Markers

Source: .claude/skills/shared/sync-inline-versions.md

SYNC:ai-sdd-artifact-contract

AI-SDD Artifact Contract — Shared spec-driven development rules stay portable and source-owned.

1. Keep reusable AI-SDD principles in .claude; put repository-specific paths, commands, owners, products, and formats in project config/reference docs.
2. Preserve cycle: spec -> plan -> tasks -> implement -> verify -> update spec/docs.
3. Trace every requirement or invariant through decision, task, TC/test, source evidence, and docs/spec update.
4. Treat code-to-spec extraction as reference-only until accepted by the canonical spec owner.
5. Any supported AI tool may plan, implement, review, or verify with synced context; using multiple tools is optional.
6. Update .claude source first, then sync generated mirrors; do not manually edit .agents, .codex, or AGENTS.md. — why: mirrors are generated artifacts; hand-edits are overwritten on the next sync
7. If docs/project-config.json, root instruction files, or a required project-reference doc is missing or stale, auto-run $project-init or the narrow lower-level route before ordinary project-specific work.

Active reference: shared/sdd-artifact-contract.md in the active skills root.

SYNC:ai-sdd-artifact-contract:reminder

• MANDATORY Apply shared/sdd-artifact-contract.md; keep reusable AI-SDD in .claude and local rules in project docs.
• MANDATORY Code-to-spec extraction is reference-only until canonical acceptance; any supported AI tool may execute with synced context.
• MANDATORY Update .claude source before syncing generated mirrors; do not manually edit .agents, .codex, or AGENTS.md.
• MANDATORY Missing or stale project config, root instruction files, or required reference docs route project-specific work through $project-init or the narrow setup route automatically. [TASK-PLANNING] [MANDATORY] BEFORE executing any workflow or skill step, create/update task tracking for all planned steps, then keep it synchronized as each step starts/completes.

[LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.

Break work into small tasks (task tracking) before starting. Add final task: "Analyze AI mistakes & lessons learned".

Extract lessons — ROOT CAUSE ONLY, not symptom fixes:

1. Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
2. Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
3. Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
4. Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
5. Recurrence gate: "Would this recur in future session WITHOUT this reminder?" — No → skip $learn.
6. Auto-fix gate: "Could $code-review/$code-simplifier/$security-review/$lint catch this?" — Yes → improve review skill instead.
7. BOTH gates pass → ask user to run $learn. [CRITICAL-THINKING-MINDSET] Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination principle: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination. AI Attention principle (Primacy-Recency): Put the 3 most critical rules at both top and bottom of long prompts/protocols so instruction adherence survives long context windows. Goal-driven execution: Define success criteria first, loop until verified, and stop only when observable checks pass. Tests verify intent: Tests must protect business rules/invariants and fail when the protected intent breaks, not only mirror current behavior.

Common AI Mistake Prevention (System Lessons)

• Re-read files after context compaction. Edit requires prior Read in same context; compaction wipes read state. Re-read before editing.
• Grep for old terms after bulk replacements. AI over-trusts find/replace completeness. Grep full repo after bulk edits for missed refs in docs/configs/catalogs.
• Check downstream references before deleting. Deletions cascade doc/code staleness. Map referencing files before removal.
• After memory loss, check existing state before creating new. Compaction wipes prior-work memory. Query current state to resume — never blindly duplicate.
• Verify AI-generated content against actual code. AI hallucinates APIs, class names, method signatures. Grep to confirm existence before documenting/referencing.
• Trace full dependency chain after edits. Changing a definition misses downstream consumers. Trace the full chain.
• When renaming, grep ALL consumer file types. Some file types silently ignore missing refs (no compile error). Search code, templates, configs, generated files.
• Trace ALL code paths when verifying correctness. Code existing ≠ code executing. Trace early exits, error branches, conditional skips — not just happy path.
• Update docs that embed canonical data when source changes. Docs inlining derived data (workflows, schemas, configs) go stale silently. Update all embedding docs alongside source.
• Verify sub-agent results after context recovery. Background agents may finish while parent compacted — grep-verify output, don't trust assumed completion.
• Cross-check full target list against sub-agent assignments. Parallel sub-agents by category miss boundary items. Reconcile union of assignments against target list before proceeding.
• Sub-agents inherit knowledge only from their agent .md definition — use custom agent types, not built-in Explore. Tool adoption = permission + knowledge + enforcement (numbered workflow step).
• Persist sub-agent findings incrementally, not as a final batch. Long sub-agents hit cutoffs before final write — findings lost. Instruct append-per-section to report file.
• When debugging, ask "whose responsibility?" before fixing. Trace caller (wrong data) vs callee (wrong handling). Fix at responsible layer — never patch symptom site.
• Test failure → record a provisional verdict before trace/edit, then investigate. Use the full five-way taxonomy: SOURCE-WRONG (production violates intent), TEST-WRONG (assertion/setup is stale), TEST-NOT-OPTIMAL (valid but fragile or low-signal test), ENVIRONMENT-BLOCKED (external state prevents a verdict), or AMBIGUOUS (intent/evidence cannot choose safely). Then trace root cause and triangulate against the governing spec (docs/specs/** if one exists) AND source. NEVER weaken an assertion, add a skip, relax a timeout, or change source merely to force green.
• Grep ALL removed names after extraction/refactoring. Primary file "done" ≠ secondary files clean. Grep entire scope for every removed symbol before declaring complete.
• Assume existing values are intentional — ask WHY before changing OR flagging one as a defect. Pattern-matching as "wrong" skips context. Before changing or reporting any constant/limit/flag/cutoff: read comments, git blame, the CALLER's ordering (the guarantee that makes the value correct usually lives in code running immediately BEFORE the cited line), and 2+ sibling call sites of the same convention. A doc stating WHAT without WHY is missing rationale, not proof of a missing guard — and in a validation pass, an accurate file:line citation proves the transcription, never the defect.
• Verify ALL affected outputs, not just the first. One build green ≠ all green. Multi-stack changes (backend/frontend/tests/docs) require verifying EVERY output.
• Evaluate fit before copying a nearby pattern. Closest example ≠ matching preconditions — verify the new context shares the same constraints, base classes, scope, lifetime.
• Holistic-first debugging — resist nearest-attention trap. Don't dive into first plausible cause. List EVERY precondition (config, env vars, paths, DB, endpoints, creds, versions, DI, data). Verify each against evidence (grep/query — not reasoning). Ask "what would falsify this?" — if nothing, it's not a hypothesis. Most expensive failure: going deeper in "obvious" layer while bug sits in layer never questioned.
• Surgical changes — apply the diff test (context-aware). Two modes: (1) Bug fix → every line traces to the bug; no restyling; orphan cleanup only for imports YOUR changes made unused. (2) Review/enhancement → implement improvements AND announce as "Enhancement beyond main request: [what]". Never silently scope-creep. Diff test: "Would this line exist if I wasn't asked to do X?" — if no, delete or announce.
• Surface ambiguity before coding — don't pick silently. Multiple valid interpretations → present each with effort: "[Request] could mean (1) [N h], (2) [N h]. Which matters?" List scope/format/volume/constraints assumptions first. If simpler path exists, say so. Never silently pick.
• [MANDATORY FIRST ACTION] ALWAYS activate a suitable skill or workflow BEFORE responding. Match task against workflow catalog + skill list; invoke via skill invocation or $start-workflow <workflowId>. NEVER answer or write code before checking. Skip = protocol violation.
• Why-Review adversarial mindset — apply when reviewing any plan, decision, or design. Default SKEPTIC not VALIDATOR: steel-man a rejected alternative, invert each stated reason ("what does it sacrifice?"), stress-test top 2-3 assumptions, run pre-mortem ("ships, fails in 3 months — what breaks?"), surface 1-2 alternatives author missed. Section presence ≠ quality; quality = causal reasoning + concrete mitigations + evidence, not "it's better" or "monitor closely".
• Front-load report-write in sub-agent prompts for large reviews. Many-file sub-agents hit budget before final write — findings lost. Design prompts so: (1) report-write is first explicit deliverable, (2) append per-file/section (not batched), (3) scope bounded so reads don't exhaust budget. Truncated mid-sentence with no report file → spawn narrower scope, don't retry same prompt.
• After context compaction, re-verify all prior phase outcomes before continuing. Summaries describe intent, not environment state (git index, filesystem, processes). On resume, FIRST audit: git status, re-read modified files, verify filesystem. Every "completed" claim is an untested hypothesis until evidence confirms.
• OOM/memory: check row count before row size. Triage: (1) Unbounded query — no DB filter for trigger? Push filter to DB; eliminates OOM. (2) Large rows? Projection reduces proportionally. Row reduction > projection in ROI.
• Assert the outcome your system OWNS, never the intermediate state your INFRASTRUCTURE owns. When testing anything asynchronous (queue/broker delivery, retries, background jobs, caches, replication), assert the final business/entity state. NEVER assert the delivery bookkeeping — consume/send status, attempt counts, last-error, row existence or counts in a broker, scheduler, or outbox/inbox table. That bookkeeping lives in shared infrastructure that ANY co-running process (a peer worker, a second replica, a leftover local container) can write, usually under a deterministic shared key, so the assertion silently tests the developer's environment instead of the system: green when run alone, flaky the instant anything else shares that broker + database. Gate question for every assertion: "would this hold no matter WHICH process did the work?" — if no, assert the converged data state instead. Corollary: process-local fault injection and in

…(truncated)