Skip to content
cccskills
FreePublished on GitHub

recall

Use when asked to recover prior work and current status before resuming. Not for handoff snapshots: use handoff. Not for source or remote-system changes.

The English summary is being prepared. Showing the original description.

See how to install

Files included (2)

  • SKILL.md3.4 KB
  • agents/openai.yaml132 B

SKILL.md (original)

Review the instructions the agent will receive before you install.

Recall

Contract

FieldBound contract
TriggerRecover prior work and current status before resuming.
AuthorityRead-only. No file, VCS, credential, paid, published, deployed, or remote mutation.
Side effectChat-output only. No file, credential, paid, published, deployed, or remote mutation.
DoneStatus-tagged capsule and one concrete next move.

Refusals

  • Fabricating state: rejected. If no session file is found, return session-not-found. Do not invent artifacts or actions.
  • Suppressing divergences: rejected. Verified records with divergence flags are returned; divergences are not hidden to satisfy the done predicate.
  • Source or remote-system changes: rejected. This skill is read-only throughout.

Inputs

  • Session identifier (required): the active session ID or session directory path to recover.
  • Current working directory (required): the directory the agent was operating in when the session ended.
  • Optional scope hint: a specific file, module, or task tag to narrow recovery. Omit to recover the full capsule.

Procedure

  1. Identify the session source. Locate the session state file or transcript corresponding to the supplied session identifier. If no session file is found, return session-not-found. Done when: the session file is located or session-not-found is returned.
  2. Extract the transcript. Parse the session transcript. If parsing fails, return transcript-parse-failure. Done when: the transcript is parsed or transcript-parse-failure is returned.
  3. Mine the transcript for work state. Collect all completed actions, emitted artifacts, tool results, and navigation events. Extract each as a discrete record with its outcome. Done when: every work record is extracted with its outcome.
  4. Verify against live repository state. For each file, artifact, and VCS-tracked target mentioned in the transcript, read the live filesystem and confirm the record matches current state. Flag each mismatch as a divergence. Done when: every record is verified or flagged as divergent.
  5. Tag the capsule. Assign each recovered record one of: completed, in-progress, blocked, or diverged. Done when: every record has a status tag.
  6. Identify the concrete next move. From the in-progress and blocked records, select the highest-priority actionable item that is not blocked by an unresolved divergence. State it as one concrete next move. Done when: one concrete next move is stated.
  7. Emit the capsule. Return a structured capsule containing all tagged records, the concrete next move, and a summary of divergences. Done when: the capsule is returned.

Failure and recovery

  • session-not-found: return with the identifier. Do not fabricate state.
  • transcript-parse-failure: return with the parse error. Do not continue with partial data.
  • no-recoverable-state: return when the transcript contains no identifiable work records. Do not invent artifacts or actions.

If some records verify and others diverge, return the verified records with divergence flags. This skill performs no mutations; no rollback is required.

Output

A structured capsule with session_id, records (each with action, outcome, status, divergence_note), next_move, and divergences, ordered as listed.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Use when explaining System V AMD64, ARM AAPCS, RISC-V psABI, stack frames, variadic calls, or FFI register rules. Not for the Rust FFI binding layer: use rust-ffi.

The English summary is being prepared. Showing the original description.

OutlineDriven/odin-claude-plugin38Updated Sep 29, 2026

Use when configuring ADC sampling time, DMA-driven ADC, calibration, or DAC channel setup on bare-metal MCUs. Not for the DMA stream itself: use dma-baremetal.

The English summary is being prepared. Showing the original description.

OutlineDriven/odin-claude-plugin38Updated Sep 29, 2026

af-xdp

Free

Use when creating AF_XDP sockets, configuring UMEM and XSK rings, writing an XDP redirect program, or choosing copy versus zero-copy mode. Not for full kernel bypass: use dpdk.

The English summary is being prepared. Showing the original description.

OutlineDriven/odin-claude-plugin38Updated Sep 29, 2026

Use when a completed session needs an agent-environment retrospective. Not for an engineering retrospective from telemetry: use engineering-retrospective.

The English summary is being prepared. Showing the original description.

OutlineDriven/odin-claude-plugin38Updated Sep 29, 2026

Use when a redacted, trimmed agent transcript must be appended to a GitHub PR or issue body, with human approval and preview. Not for automated or model-initiated insertion.

The English summary is being prepared. Showing the original description.

OutlineDriven/odin-claude-plugin38Updated Sep 29, 2026

Use when a repo needs agent setup, AGENTS.md added or made lean, CLAUDE.md audited, or agent instructions scored or pruned. Not for remote, credential, publish, deploy, or irreversible changes.

The English summary is being prepared. Showing the original description.

OutlineDriven/odin-claude-plugin38Updated Sep 29, 2026

See all skills by OutlineDriven

Report a problem with this skill