Long Task Continuation
Overview
Keep long work checkpointed, resumable, drift-aware, and evidence-gated. This
protocol does not execute plans, dispatch subagents, run tests, or grant
completion authority.
Authority Boundary
The Method Pack owns continuation discipline only. It does not own the parent
plan, host retry/watchdog behavior, authoritative GateDecision, evidence
sufficiency, requirement acceptance, or completion.
When To Use
Use this skill when the work has meaningful phases, may be compacted/resumed or
handed off, uses subagents, or explicitly needs continuity and drift control.
Architecture, contract, shared-workflow, and verification-gate changes also
benefit from it. Do not force it onto a short answer or one-command check.
Choose exactly one state carrier:
- within this continuation workflow, use a durable
work/ record for
medium/high work, or for work of any complexity that actually crosses
sessions, needs handoff, or requires resumable state;
- otherwise keep one inline checkpoint.
Multi-step, todo-driven, possible-compaction, and subagent use do not force
durable records for low-complexity work by themselves. Do not create both carriers or a record per
slice.
Required Artifacts
A durable task has one process trail under
docs/aegis/work/YYYY-MM-DD-<slug>/. It keeps logical intent/baseline state,
the latest todo/checkpoint/resume state, terminal evidence/drift state, and a
completion reflection when warranted. These are
TaskIntentDraft, BaselineReadSetHint, BaselineUsageDraft,
ImpactStatementDraft, TodoCheckpointDraft, ResumeStateHint,
DriftCheckDraft, and EvidenceBundleDraft views—not authoritative runtime
records or separate plan owners.
Read only the lifecycle-matched section of durable-work-guidance.md:
## Required Artifact Layout and ## Create A Durable Work Record for a new durable work record;
## Update A Slice when an existing helper-backed record needs sidecar updates;
## Retry Convergence Detail for retry/attempt bookkeeping;
## Pause, Handoff, And Completion Bundle when preparing a pause, handoff, or completion bundle; and
## Expanded State Fields only when natural checkpoint prose is ambiguous.
The reference owns artifact layout and <aegis-workspace-helper> command
detail; this file owns carrier selection, resume order, drift decisions, and
stop conditions.
An Execution Readiness View may be kept in the intent or active checkpoint
for medium/high, handoff-prone, long-running, subagent-driven, architecture,
contract, compatibility, or retirement-sensitive work. It renders existing
intent, scope, baseline, owner, test, review, and drift constraints; it is not a
new JSON artifact or completion authority.
Planless Slice Lane:
- When an existing parent plan/spec owns a bounded task, reuse it and the
current checkpoint. For a no-parent direct bounded request with no new
durable or unclear verification boundary, use an inline checkpoint.
- State one compact
Slice Card: Goal, Parent plan/spec (or
none — direct bounded request), Files, Boundary, Verification, and Stop.
- The slice goal closes only that slice. Final completion returns to the parent
or direct bounded request through
verification-before-completion.
- Do not create a plan/spec merely to give a micro-slice a parent, and do not
create per-slice plans/specs or work records.
- Escalate when a new owner, contract, schema, public API, architecture,
migration, persistence, security/permission, distribution/release surface,
unclear verification boundary, or mismatch with parent scope or acceptance
appears.
When durable architecture decisions are in scope, these work records are the
preferred ADR Auto Backfill source. Preserve decision signals, source refs,
alternatives, compatibility, retirement, drift, and baseline-sync questions.
Start Protocol
Before execution:
- Capture requested outcome, scope, non-goals, risks, parent plan/goal, success
evidence, and stop states (
done | blocked | needs-verification | scope-exceeded).
- Identify required baseline refs and record acknowledged, cited, and missing
refs. Missing authority pauses in
needs-baseline-readback.
- Choose inline or durable state once, then record the todo map, active slice,
completed slices/evidence, blockers, next step, and current branch/HEAD.
- When an
Execution Readiness View exists, retain its intent lock, scope
fence, baseline lock, compatibility/retirement boundary, tests, reviews,
evidence, and rewind rules.
- For a new helper-backed record, use
durable-work-guidance.md to create and
structurally check it before implementation.
Retry Convergence Protocol
A failed verification is another attempt in the current slice, not a new
slice. Keep failed-attempt telemetry out of terminal evidence and normal
commits. Only evidence-finalized, blocked, or abandoned is terminal.
When retry state reaches process-artifact-pressure, stop auto-retry and route
to systematic-debugging or verification-before-completion. Load the durable
reference only when the attempt/evidence commands or sidecar rules are needed.
Per-Slice Protocol
Before each slice, state the current goal/todo, intended edits, explicit
non-edits, verification, and readiness alignment. A bounded parent-plan or
no-parent slice uses the compact Slice Card rather than a new plan/spec.
After each slice, update completed todos, evidence refs, newly used baseline
refs, blockers, next step, and drift decision. When an active helper-backed work record exists, read durable-work-guidance.md and update that same record;
never create another workstream for bookkeeping.
When patch-shape/ripple triage, an H-class finding, or a bounded compatibility
mitigation fired, a locally green result does not clear the direction. Retain
PatchShape, CanonicalOwner, UpwardDrillSignal, latest outcome, and one
bounded evidence ref; do not copy raw logs or full diffs. If no fresh evidence
exists, the state is needs-verification or partial. For repairs or
optimization, retain the causal scope, evidence references, unresolved
influences that could change the solution, and verification coverage in the
same checkpoint. Compression must not turn one confirmed cause into a complete
root claim or erase counterevidence; recheck these unknowns before resuming
retained changes. Do not add a parallel causal record.
Resume Protocol
Resume in this order:
- Read original intent, parent plan/goal, latest checkpoint and resume hint.
Reconcile current user instructions: continuation, amendment, pause, or
replacement. Preserve outstanding deliverables; update authorized scope
before treating a valid amendment as drift. Internal milestones are not
substitutes for the user's delivered result.
- Re-read required baseline refs and relevant active
CONTEXT.md language.
- Read the
Execution Readiness View when present.
- Compare checkpoint branch/HEAD, completed commits, evidence refs, and claims
with the current worktree.
Treat inherited completion claims as unverified until the worktree confirms
them. Correct contradictions in the checkpoint, disclose them in the final
report, and do not republish stale claims as current.
Preserve the claim's subject, scope and resume-time state; distinguish that
from later repair and from historical execution you cannot verify. One
accurate correction may cover related claims. The record must adopt it,
without elsewhere denying it or assigning it to another scope. A reviewer's
classification is contestable: if material disagreement remains unresolved
after checking evidence, retain
needs-verification without forcing assent.
If evidence resolves it, retain the supported conclusion and verify normally.
- Compare the active slice against intent lock, scope fence, baseline lock,
compatibility/retirement boundary, tests, reviews, and non-goals.
- Re-run the drift decision, then name the next smallest authorized action.
Any disagreement among plan, checkpoint, baseline, context, readiness view, or
worktree pauses execution. A semantic conflict routes to
establishing-project-context; an unplanned repair re-reads the retained
invariant, owner seam, patch shape, and causal topology, then route comparison to
systematic-debugging. A new carrier name alone does not prove a new direction. Never resume from memory alone.
Drift Check
Check original intent and stop condition, parent scope/acceptance, compatibility,
new owners/fallbacks/adapters/branches, retirement, evidence freshness, and any
readiness locks. Allowed decisions are continue, pause-for-user,
needs-baseline-readback, needs-verification, and blocked.
Never emit gate-passed, completion-granted, or authoritatively-safe.
Completion Candidate Protocol
Before a completion claim:
- Use
aegis:verification-before-completion.
- Confirm every todo has status, blockers are resolved/externalized, evidence
covers acceptance, and drift has no blocking state.
- If a durable record exists, load
durable-work-guidance.md for the completion
bundle and structural workspace check.
- For durable architecture work, pass the work record, proof bundle and ADR signals
to verification for ADR Backfill Check.
Generated packs are future-runtime inputs only. Method Pack output remains
verified evidence and advisory judgment, not authoritative completion.
Minimal Reporting Shape
Report naturally and omit empty structures. Keep these semantic slots visible:
Aegis Visibility; current todo/active/completed/next; baseline usage decision;
readiness state when present; fresh evidence; inherited-claim corrections when
a resumed record overstated its state; retry/convergence state when
relevant; drift decision; risk/unknown; and the next smallest safe action.
1---2name: long-task-continuation3description: "Use when a task is multi-step, may span context resets or sessions, uses subagents, or risks losing state before completion."4metadata:5 source: https://skillmd.com/skills/hashgraph-online/long-task-continuation6---78# Long Task Continuation910## Overview1112Keep long work checkpointed, resumable, drift-aware, and evidence-gated. This13protocol does not execute plans, dispatch subagents, run tests, or grant14completion authority.1516## Authority Boundary1718The Method Pack owns continuation discipline only. It does not own the parent19plan, host retry/watchdog behavior, authoritative `GateDecision`, evidence20sufficiency, requirement acceptance, or completion.2122## When To Use2324Use this skill when the work has meaningful phases, may be compacted/resumed or25handed off, uses subagents, or explicitly needs continuity and drift control.26Architecture, contract, shared-workflow, and verification-gate changes also27benefit from it. Do not force it onto a short answer or one-command check.2829Choose exactly one state carrier:3031- within this continuation workflow, use a durable `work/` record for32 medium/high work, or for work of any complexity that actually crosses33 sessions, needs handoff, or requires resumable state;34- otherwise keep one inline checkpoint.3536Multi-step, todo-driven, possible-compaction, and subagent use do not force37durable records for low-complexity work by themselves. Do not create both carriers or a record per38slice.3940## Required Artifacts4142A durable task has one process trail under43`docs/aegis/work/YYYY-MM-DD-<slug>/`. It keeps logical intent/baseline state,44the latest todo/checkpoint/resume state, terminal evidence/drift state, and a45completion reflection when warranted. These are46`TaskIntentDraft`, `BaselineReadSetHint`, `BaselineUsageDraft`,47`ImpactStatementDraft`, `TodoCheckpointDraft`, `ResumeStateHint`,48`DriftCheckDraft`, and `EvidenceBundleDraft` views—not authoritative runtime49records or separate plan owners.5051Read only the lifecycle-matched section of `durable-work-guidance.md`:5253- `## Required Artifact Layout` and `## Create A Durable Work Record` for a new durable work record;54- `## Update A Slice` when an existing helper-backed record needs sidecar updates;55- `## Retry Convergence Detail` for retry/attempt bookkeeping;56- `## Pause, Handoff, And Completion Bundle` when preparing a pause, handoff, or completion bundle; and57- `## Expanded State Fields` only when natural checkpoint prose is ambiguous.5859The reference owns artifact layout and `<aegis-workspace-helper>` command60detail; this file owns carrier selection, resume order, drift decisions, and61stop conditions.6263An `Execution Readiness View` may be kept in the intent or active checkpoint64for medium/high, handoff-prone, long-running, subagent-driven, architecture,65contract, compatibility, or retirement-sensitive work. It renders existing66intent, scope, baseline, owner, test, review, and drift constraints; it is not a67new JSON artifact or completion authority.6869Planless Slice Lane:7071- When an existing parent plan/spec owns a bounded task, reuse it and the72 current checkpoint. For a no-parent direct bounded request with no new73 durable or unclear verification boundary, use an inline checkpoint.74- State one compact `Slice Card`: Goal, `Parent plan/spec` (or75 `none — direct bounded request`), Files, Boundary, Verification, and Stop.76- The slice goal closes only that slice. Final completion returns to the parent77 or direct bounded request through `verification-before-completion`.78- Do not create a plan/spec merely to give a micro-slice a parent, and do not79 create per-slice plans/specs or work records.80- Escalate when a new owner, contract, schema, public API, architecture,81 migration, persistence, security/permission, distribution/release surface,82 unclear verification boundary, or mismatch with parent scope or acceptance83 appears.8485When durable architecture decisions are in scope, these work records are the86preferred ADR Auto Backfill source. Preserve decision signals, source refs,87alternatives, compatibility, retirement, drift, and baseline-sync questions.8889## Start Protocol9091Before execution:92931. Capture requested outcome, scope, non-goals, risks, parent plan/goal, success94 evidence, and stop states (`done | blocked | needs-verification |95 scope-exceeded`).962. Identify required baseline refs and record acknowledged, cited, and missing97 refs. Missing authority pauses in `needs-baseline-readback`.983. Choose inline or durable state once, then record the todo map, active slice,99 completed slices/evidence, blockers, next step, and current branch/HEAD.1004. When an `Execution Readiness View` exists, retain its intent lock, scope101 fence, baseline lock, compatibility/retirement boundary, tests, reviews,102 evidence, and rewind rules.1035. For a new helper-backed record, use `durable-work-guidance.md` to create and104 structurally check it before implementation.105106## Retry Convergence Protocol107108A failed verification is another attempt in the current slice, not a new109slice. Keep failed-attempt telemetry out of terminal evidence and normal110commits. Only `evidence-finalized`, `blocked`, or `abandoned` is terminal.111When retry state reaches `process-artifact-pressure`, stop auto-retry and route112to `systematic-debugging` or `verification-before-completion`. Load the durable113reference only when the attempt/evidence commands or sidecar rules are needed.114115## Per-Slice Protocol116117Before each slice, state the current goal/todo, intended edits, explicit118non-edits, verification, and readiness alignment. A bounded parent-plan or119no-parent slice uses the compact Slice Card rather than a new plan/spec.120121After each slice, update completed todos, evidence refs, newly used baseline122refs, blockers, next step, and drift decision. When an active helper-backed work record exists, read `durable-work-guidance.md` and update that same record;123never create another workstream for bookkeeping.124125When patch-shape/ripple triage, an H-class finding, or a bounded compatibility126mitigation fired, a locally green result does not clear the direction. Retain127`PatchShape`, `CanonicalOwner`, `UpwardDrillSignal`, latest outcome, and one128bounded evidence ref; do not copy raw logs or full diffs. If no fresh evidence129exists, the state is `needs-verification` or `partial`. For repairs or130optimization, retain the causal scope, evidence references, unresolved131influences that could change the solution, and verification coverage in the132same checkpoint. Compression must not turn one confirmed cause into a complete133root claim or erase counterevidence; recheck these unknowns before resuming134retained changes. Do not add a parallel causal record.135136## Resume Protocol137138Resume in this order:1391401. Read original intent, parent plan/goal, latest checkpoint and resume hint.141 Reconcile current user instructions: continuation, amendment, pause, or142 replacement. Preserve outstanding deliverables; update authorized scope143 before treating a valid amendment as drift. Internal milestones are not144 substitutes for the user's delivered result.1452. Re-read required baseline refs and relevant active `CONTEXT.md` language.1463. Read the `Execution Readiness View` when present.1474. Compare checkpoint branch/HEAD, completed commits, evidence refs, and claims148 with the current worktree.149 Treat inherited completion claims as unverified until the worktree confirms150 them. Correct contradictions in the checkpoint, disclose them in the final151 report, and do not republish stale claims as current.152 Preserve the claim's subject, scope and resume-time state; distinguish that153 from later repair and from historical execution you cannot verify. One154 accurate correction may cover related claims. The record must adopt it,155 without elsewhere denying it or assigning it to another scope. A reviewer's156 classification is contestable: if material disagreement remains unresolved157 after checking evidence, retain `needs-verification` without forcing assent.158 If evidence resolves it, retain the supported conclusion and verify normally.1595. Compare the active slice against intent lock, scope fence, baseline lock,160 compatibility/retirement boundary, tests, reviews, and non-goals.1616. Re-run the drift decision, then name the next smallest authorized action.162163Any disagreement among plan, checkpoint, baseline, context, readiness view, or164worktree pauses execution. A semantic conflict routes to165`establishing-project-context`; an unplanned repair re-reads the retained166invariant, owner seam, patch shape, and causal topology, then route comparison to167`systematic-debugging`. A new carrier name alone does not prove a new direction. Never resume from memory alone.168169## Drift Check170171Check original intent and stop condition, parent scope/acceptance, compatibility,172new owners/fallbacks/adapters/branches, retirement, evidence freshness, and any173readiness locks. Allowed decisions are `continue`, `pause-for-user`,174`needs-baseline-readback`, `needs-verification`, and `blocked`.175176Never emit `gate-passed`, `completion-granted`, or `authoritatively-safe`.177178## Completion Candidate Protocol179180Before a completion claim:1811821. Use `aegis:verification-before-completion`.1832. Confirm every todo has status, blockers are resolved/externalized, evidence184 covers acceptance, and drift has no blocking state.1853. If a durable record exists, load `durable-work-guidance.md` for the completion186 bundle and structural workspace check.1874. For durable architecture work, pass the work record, proof bundle and ADR signals188 to verification for ADR Backfill Check.189190Generated packs are future-runtime inputs only. Method Pack output remains191verified evidence and advisory judgment, not authoritative completion.192193## Minimal Reporting Shape194195Report naturally and omit empty structures. Keep these semantic slots visible:196`Aegis Visibility`; current todo/active/completed/next; baseline usage decision;197readiness state when present; fresh evidence; inherited-claim corrections when198a resumed record overstated its state; retry/convergence state when199relevant; drift decision; risk/unknown; and the next smallest safe action.