Index of every page on this site: llms.txt
All skills
wix avatar

/rp-user-action-recovery

@e2bc48e official
by Wix.comwix/skills40 stars
42
Run it yourself

Regenerates a stale durable human request without accepting an outdated answer.

  • 1 file
  • 2.5 KB
  • Updated last week
  • GitHub
Use this Skill: https://skilld.dev/gh/wix/skills/rp-user-action-recovery

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈26 tokens for metadata: the name and description. ≈611 when used: this file.

Description uses 1.3% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈105 characters, or 1.3%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

rp-user-action-recovery

Use when the router reports user_action_stale. The request's pinned evidence changed or disappeared, so answering it is unsafe. Recover only the reported requestIds.

  1. Read each request through scripts/user-action.js show --project <projectDir> --request <id>. Identify its stage, effect and pinned artifacts. Do not read secret-bearing config into tool output. Preserve the existing request and all recorded answers.
  2. For effect read-back-omission-ack, run node scripts/user-action.js recover-omission-ack --project <projectDir> --request <stale-id>. Its producer rebuilds the evidence from approved omissions, checkpoints the new request and ledger bindings, then supersedes the old request. Retry the same command after an interruption. Changed mapping/import inputs require revalidation by their producing stage; do not refresh their hashes. Regenerate the completion report so its operator action uses the new request, then route again for the human answer. Skip the generic steps below.
  3. Use the producing stage's contract to rebuild any missing review evidence and prepare a fresh explanation and choices for the current evidence. Do not merely change the old digest, reuse outdated prose, answer on the user's behalf, or perform setup/import writes. If the evidence cannot be regenerated, report the specific repair needed and stop.
  4. Create a new pending request with the same stage, kind and effect scope through scripts/user-action.js create --project <projectDir> --request-file <json>. Pin the current reviewed artifacts. An evidence digest may change; the subject and permission being requested must remain the same. A materially different decision needs its own producer-owned recovery rather than silently changing this request's scope.
  5. Only after the replacement exists, run: node scripts/user-action.js supersede --project <projectDir> --request <stale-id> --replacement <fresh-id>. This refuses a missing, answered, stale or differently scoped replacement. No approval effect is applied. Route again and present the fresh request for a human answer.

Output

The old request remains in orchestration/user-actions/ as superseded; a fresh pending replacement carries the current review and artifact hashes. Approvals remain unchanged. If recovery is blocked, preserve the old request and identify the required repair.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at e2bc48e. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated last week

README badge

README badge for wix/skills/rp-user-action-recovery