Spec Kitty Runtime Next

spec-kitty-runtime-next

This skill teaches agents how to advance a Spec Kitty mission through the canonical runtime control loop, including doctrine-aware context loading at each step boundary.

When to Use This Skill

Use this skill when the user wants to:

• Advance a mission to its next step
• Understand what the runtime will do next
• Unblock a stalled mission
• Interpret runtime outcomes (step, blocked, decision_required, terminal)

How the Runtime-Next System Works

The spec-kitty next command is the single entry point for agent-driven mission execution. Each call returns a deterministic decision about what action the agent should take next.

Decision Algorithm

The runtime evaluates state in this order:

1. Mission state machine — Current phase and available transitions (from mission-runtime.yaml DAG)
2. WP iteration check — For implement and review steps, the CLI bridge manages WP-level iteration WITHOUT advancing the runtime. The runtime only advances when ALL WPs reach terminal/handoff lanes.
3. Guard conditions — Required artifacts, prerequisites, dependency graph
4. Priority ordering — Reviews before implementations, higher-priority WPs first, dependency-free WPs before dependent ones

WP Iteration Logic (Critical)

The CLI bridge (not the runtime) manages WP-level iteration:

• If current step is implement or review
• AND there are WPs in planned or in_progress lanes
• THEN return a WP-level decision without advancing the runtime step
• The runtime step only advances when ALL WPs are in terminal/handoff lanes (done, approved, or for_review)

This means multiple calls to spec-kitty next during implementation will return different WP IDs but the same step_id (e.g., "implement") until all WPs are done.

Mission Runtime YAML Schema

Missions define steps as a DAG (directed acyclic graph) with dependencies:

mission:
  key: software-dev
  name: Software Dev Kitty
  version: "2.1.0"

steps:
  - id: discovery
    title: Discovery & Research
    depends_on: []
    prompt_template: research.md

  - id: specify
    depends_on: [discovery]
    prompt_template: specify.md

  - id: plan
    depends_on: [specify]
    prompt_template: plan.md

  - id: tasks
    depends_on: [plan]

  - id: implement
    depends_on: [tasks]
    prompt_template: implement.md

  - id: review
    depends_on: [implement]
    prompt_template: review.md

  - id: accept
    depends_on: [review]
    prompt_template: accept.md

Decision Kinds

Every call to spec-kitty next returns exactly one decision kind:

Decision Output Fields

{
  "kind": "step",
  "agent": "claude",
  "mission_slug": "042-test-mission",
  "mission": "software-dev",
  "mission_state": "implementing",
  "action": "implement",
  "wp_id": "WP02",
  "workspace_path": ".worktrees/042-test-mission-lane-b",
  "prompt_file": "/tmp/spec-kitty-next-claude-042-test-mission-implement-WP02.md",
  "reason": null,
  "guard_failures": [],
  "progress": {
    "total_wps": 5,
    "done_wps": 1,
    "approved_wps": 0,
    "in_progress_wps": 1,
    "planned_wps": 3,
    "for_review_wps": 0
  },
  "run_id": "abc123",
  "step_id": "implement",
  "decision_id": null,
  "question": null,
  "options": null
}

6 Guard Primitives

Guards block step transitions by returning failure descriptions:

Guards never raise exceptions — they return false on missing context.

Prompt File Generation

The runtime generates a temp file at: /tmp/spec-kitty-next-{agent}-{mission_slug}-{action}[-{wp_id}].md

Template actions (specify, plan, tasks): Mission context header + governance context + action-specific template content.

WP actions (implement, review): Full isolation-aware prompt containing:

1. WP header with workspace path
2. Governance context (paradigms, directives, tools)
3. WP Isolation Rules — DO only modify this WP's status, DO NOT change other WPs or react to their status changes
4. Working directory and review commands
5. WP file content (from tasks/WP##.md)
6. Completion instructions

Decision prompts: Question text, options, and the --answer command to run.

Run Persistence

Runtime state is persisted between calls:

.kittify/runtime/
├── feature-runs.json       # Index: {"mission-slug": {"run_id": "...", "run_dir": "..."}}
└── runs/
    └── <run_id>/
        └── state.json      # Runtime snapshot (current step, inputs, etc.)

Mission Detection

When --mission is omitted, the runtime detects the mission via (in order):

1. SPECIFY_MISSION environment variable
2. Git branch name (mission and lane branches both encode the mission slug)
3. Current directory path (walks up looking for ###-mission-name)
4. Single mission auto-detect (only if exactly one mission exists)
5. Error with guidance if ambiguous

NOTE: Always use --mission <slug> in multi-mission repositories.

Doctrine-Aware Step Execution

The runtime-next loop should load doctrine context iteratively — not all at once. Each step boundary is a context loading opportunity.

Agent Profile at Init

At the start of a session, resolve the active agent profile. This scopes your role, boundaries, and initialization context.

Load the profile using the Python API — do NOT read YAML files directly:

from charter.activation.doctrine_service_builder import build_activation_aware_doctrine_service

service = build_activation_aware_doctrine_service(project_root)

# resolve_profile()'s specializes_from lineage traversal is a repository
# operation, not available on the filtered `agent_profiles` dict — reach it
# through the pinned lineage/mutation accessor:
repo = service.agent_profile_repository
profile = repo.resolve_profile("<profile-id>")  # e.g. "implementer"

# Internalize identity — acknowledge this at session start
print(profile.initialization_declaration)

# Respect scope boundaries
profile.specialization.primary_focus       # What you actively do
profile.specialization.avoidance_boundary  # What you must NOT do
profile.collaboration.handoff_to           # Roles to defer to when out of scope

# Load only the directives this profile references (same `service`, gated dict)
for ref in profile.directive_references:
    directive = service.directives.get(f"DIRECTIVE_{ref.code}")

Discovery (if you don't know your profile-id):

spec-kitty agent profile list
spec-kitty agent profile show <profile-id>

Action-Scoped Context at Each Step

At each step boundary (when spec-kitty next returns a step decision), load governance context scoped to the current action — not the full doctrine:

# Load only what's relevant to this action (compact after first load)
spec-kitty charter context --action implement --json

The context system uses two depth levels:

First-load state is tracked per action in .kittify/charter/context-state.json. This means implement and review each get their own first-load bootstrap independently.

Pull Specific Doctrine On Demand

When you need governance guidance mid-step (e.g., how to structure tests, which review criteria apply), pull the specific tactic or directive by ID rather than re-loading the full context:

from charter.activation.doctrine_service_builder import build_activation_aware_doctrine_service

service = build_activation_aware_doctrine_service(project_root)

# Pull a specific tactic when it becomes relevant
tactic = service.tactics.get("tdd-red-green-refactor")

# Pull a specific directive
directive = service.directives.get("TEST_FIRST")

The action index (actions/<action>/index.yaml) tells you which doctrine artifacts are relevant to the current step. Load the index to discover what to pull:

from charter.offering.missions.action_index import load_action_index

index = load_action_index(missions_root, "software-dev", "implement")
# index.directives → ["TEST_FIRST", ...]
# index.tactics → ["tdd-red-green-refactor", ...]
# index.procedures → [...]

Anti-Pattern: Upfront Context Dump

Do NOT load all doctrine into context at session start. This wastes tokens and dilutes relevance. Instead:

1. At init: Load agent profile + initialization declaration.
2. At each step boundary: Call charter context --action <action>.
3. When stuck or need guidance: Pull specific tactic/directive by ID.
4. When reviewing: Pull review-scoped doctrine, not implement-scoped.

Step 1: Load Runtime Context

Before invoking the runtime, gather the current state.

Commands:

# Check WP status for a mission
spec-kitty agent tasks status --mission <mission-slug>

# Check current context for an action
spec-kitty agent context resolve --action implement --mission <mission-slug> --json

What to look for:

• Active mission slug and mission type
• Current WP lane status (planned, claimed, in_progress, for_review, in_review, approved, done, blocked, canceled)
• Whether there are WPs ready for implementation or review
• Any blocked WPs that need attention first

Step 2: Run the Next Command

# Run the next step
spec-kitty next --agent <agent> --mission <mission-slug> --json

# After completing a step successfully
spec-kitty next --agent <agent> --mission <mission-slug> --result success --json

# After a step failed
spec-kitty next --agent <agent> --mission <mission-slug> --result failed --json

# After a step was blocked
spec-kitty next --agent <agent> --mission <mission-slug> --result blocked --json

Note: --mission is the sole selector. The legacy --feature alias has been removed (#1060); passing --feature now exits with No such option.

The --result flag tells the runtime the outcome of the previous step. If omitted, spec-kitty next returns current state without advancing (query mode). It does not report success for the previous step.

Step 3: Interpret the Result

See references/runtime-result-taxonomy.md for the complete taxonomy.

Always check guard_failures — this field may appear on any decision kind, not just blocked.

The kind="step" prompt-file contract is a hard runtime invariant (C1/C2). A kind="step" envelope MUST carry a prompt_file (or its consumer-side prompt_path alias) that is non-null, non-empty, and resolves on disk. If the runtime cannot produce an actionable step (no composed action, guard failure, blocked dependency, prompt build error, etc.), it returns kind="blocked" with a non-empty reason (and optional machine-readable code such as no_prompt_template). There is no third state: an agent loop should never observe a kind="step" decision with prompt_file == null.

Always check progress for completion. If progress.done_wps equals progress.total_wps but kind is not terminal, the mission is actually complete (known issue #335). The runtime may not detect completion when no prior run state exists. Treat this as terminal and run /spec-kitty.accept; if acceptance passes, run /spec-kitty.merge, then run mission review and the retrospective workflow.

Step 4: Handle decision_required

When the runtime needs input:

# The decision includes question, options, and decision_id
# Answer using:
spec-kitty next --agent <agent> --mission <mission-slug> \
  --result success --answer "<choice>" --decision-id "<decision_id>" --json

If the agent cannot determine the answer, escalate to the user with the question and options.

Step 5: Handle Blocked States

See references/blocked-state-recovery.md for detailed recovery patterns.

Quick diagnostic:

# Check WP status and dependency graph
spec-kitty agent tasks status --mission <mission-slug>

# Check specific WP dependencies
spec-kitty agent tasks list-dependents WP## --mission <mission-slug>

Common blockers:

Step 6: The Agent Loop

The complete agent loop pattern:

# 1. Start the loop
DECISION=$(spec-kitty next --agent claude --mission 042-mission --json)
KIND=$(echo "$DECISION" | jq -r '.kind')

# 2. Loop until terminal or unresolvable block
while [ "$KIND" = "step" ] || [ "$KIND" = "decision_required" ]; do

  # Workaround #335: check progress for completion even if kind != terminal
  DONE=$(echo "$DECISION" | jq -r '.progress.done_wps // 0')
  TOTAL=$(echo "$DECISION" | jq -r '.progress.total_wps // 0')
  if [ "$TOTAL" -gt 0 ] && [ "$DONE" -eq "$TOTAL" ]; then
    break  # Mission is actually complete
  fi

  if [ "$KIND" = "step" ]; then
    PROMPT=$(echo "$DECISION" | jq -r '.prompt_file')

    # Contract (C1/C2, post-#336 fix): kind=step always carries a
    # non-empty prompt_file resolvable on disk. If a prompt cannot be
    # resolved, the runtime emits kind=blocked with a populated reason.

    # Read and execute the prompt...
    RESULT="success"  # or "failed" or "blocked"
  elif [ "$KIND" = "decision_required" ]; then
    # Answer the question...
    RESULT="success"
  fi

  DECISION=$(spec-kitty next --agent claude --mission 042-mission --result "$RESULT" --json)
  KIND=$(echo "$DECISION" | jq -r '.kind')
done

# 3. Handle terminal state — canonical post-merge sequence
if [ "$KIND" = "terminal" ] || [ "$DONE" -eq "$TOTAL" ]; then
  # Run /spec-kitty.accept.
  # If acceptance passes, run /spec-kitty.merge.
  # After merge, follow the canonical post-merge sequence:
  #   a. Mission review: /spec-kitty-mission-review
  #   b. Author or verify retrospective:
  #      spec-kitty retrospect create --mission 042-mission  # if record absent
  #      OR verify: cat .kittify/missions/<mission_id>/retrospective.yaml
  #   c. Surface findings:
  #      spec-kitty retrospect summary                                   # read-only aggregation
  #      spec-kitty agent retrospect synthesize --mission 042-mission  # dry-run by default; --apply to mutate
  # Note: summary aggregates; synthesize applies proposals — neither authors records.
fi

The loop continues until:

• terminal — mission complete, exit loop
• query — read-only preview, no state mutation
• blocked — cannot proceed without external resolution
• decision_required — only if the agent cannot answer (escalate to user)

Important: Runtime Precedence Rules

1. Always use spec-kitty next rather than manually sequencing phases
2. Always pass --mission in multi-mission repositories
3. Respect mission state machine transitions — do not skip steps
4. Read the prompt_file — it contains the full context the agent needs
5. Check guard_failures on every decision, not just blocked ones
6. Reviews before implementations — the runtime prioritizes unblocking downstream work
7. WP isolation — only modify the WP you were assigned, ignore other WPs

Known Issues

#335 — Completed missions return step instead of terminal. When spec-kitty next is called on a mission with all WPs done but no prior runtime run state, it creates a new run starting at discovery instead of recognizing the mission is complete. Workaround: Check progress.done_wps == progress.total_wps as a secondary completion signal.

#336 — fixed. prompt_file is always non-empty and resolvable on disk on kind: step decisions. When no prompt is available, the runtime now emits a structured kind: blocked decision with a non-empty reason (and optional machine-readable code such as no_prompt_template). Agent loops no longer need to defensively null-check prompt_file; a kind: step decision with a null prompt is a runtime bug.

Standalone Invocations (Outside Missions)

Not all governed work happens inside an active mission. When a user asks for help with a task that has no active spec-kitty next loop — a code review, a quick implementation, an ad-hoc analysis — you should still invoke Spec Kitty's governance layer with standalone dispatch.

Spec Kitty never spawns a parallel LLM call. You are the host; Spec Kitty routes, assembles governance context, and records the trail.

When to use standalone dispatch

If the user says anything like "use spec kitty to ...", "hey spec kitty ...", or "spec kitty ", run standalone dispatch unless they are clearly asking for a full mission.

The governance injection loop

1. Get context:

   spec-kitty dispatch "implement the login handler" --json
   spec-kitty dispatch "review WP05" --profile reviewer --json
   

   Response includes invocation_id, governance_context_text, and governance_context_available.

2. Inject governance: Read governance_context_text and treat it as binding governance context for your task. Follow any directives and constraints it contains. If governance_context_available is false, note this to the user but proceed with the task.

3. Execute: Do the work. Generate the code, analysis, or plan.

4. Close the Op (mandatory — dispatch leaves it open):

   spec-kitty profile-invocation complete \
     --invocation-id <invocation_id> \
     --outcome <done|failed|abandoned>
   

   Use the real outcome: done for completed work, failed for work that did not succeed, abandoned for dropped work. Never leave an Op open deliberately — spec-kitty doctor ops reports and sweeps orphans.

Trail produced

Every standalone invocation writes a Tier 1 JSONL file to:

kitty-ops/<invocation_id>.jsonl

Viewable at any time with spec-kitty invocations list --json. No SaaS connection required.

For full CLI surface documentation, see src/charter/offering/skills/spec-kitty/SKILL.md.

References

• references/runtime-result-taxonomy.md -- Decision kinds, output fields, and precedence rules
• references/blocked-state-recovery.md -- 6 blocked state patterns with diagnosis and recovery