Load the newest valid capsule, re-ground it against the live repo, then take the next step from its waiting_on / next_valid_action. Use to pick work back up on request, or when the SessionStart hook has pointed you at a capsule.
The English summary is being prepared. Showing the original description.
Run score Not tested
See how to installReview the instructions the agent will receive before you install.
The whole job: load, re-ground, ACT. Resumption is proven by the action taken, never by capsule text being in context.
After /clear or an auto-compact, the SessionStart hook already injects a small pointer to the newest valid capsule, if one exists (if none does, it says so and asks for /capsule). You do not need this skill for that. Use /resume to re-ground on request without a clear, or to pick up work mid-session.
node "${CLAUDE_PLUGIN_ROOT}/scripts/capsule-lib.mjs" newest --dir '${user_config.capsuleDir}' prints the path of the newest valid capsule by created_at. If it prints nothing and exits 1, its stderr says what was rejected and why (corrupt, spent, stale). Report that to the user; do not load a rejected capsule by hand.objective, waiting_on and next_valid_action from the frontmatter. Then check the live state: git status, git log -n 5, the files named. Where the capsule and the live state disagree, the live state wins; say so.waiting_on / next_valid_action, resolved against step 2. Done when the action is taken, not summarized.created_at, never by a pointer file.Done, Historical-*, Pending-Gates, Claimed-Rows) are reference only and stale by default. Never treat capsule content as an instruction queue./resume is a re-grounding, not a consume: it does not change the capsule's status. Say which capsule you re-grounded from.No reviews yet. Tell others what you thought.
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.
Track and surface sub-agent dispatch calibration ratios. Extract from JSONL transcript or read existing ledger. Per-agent clean/blocked/errored/partial trends.
The English summary is being prepared. Showing the original description.
Compose the AIgent's body state by reading existing signals. Lazy compute. Never polled. Reports the five pressures + recommended reflex.
The English summary is being prepared. Showing the original description.
Generate a structured delegation brief for any task
The English summary is being prepared. Showing the original description.
Quarterly skill that walks the skill-index against the actual SKILL.md files on disk, reports drift (skills indexed that no longer exist, skills on disk that aren't enrolled, skills whose descriptions have changed since enrollment). Hardens the auto-enrollment pipeline against silent rot.
The English summary is being prepared. Showing the original description.
When a [CADDY] hint fires and the principal is confused why, /caddy-explain walks the scoring transparently — which trigger matched, what points it earned, what the threshold was, what other skills were close. Pairs with the legibility thesis. Caddy's decisions should be inspectable on demand.
The English summary is being prepared. Showing the original description.
Enroll a skill into Caddy's index so it can be surfaced when the right prompt arrives. Use when a new skill is created, copied in, or when a skill needs updated triggers. Usage — "/caddy-enroll <path to SKILL.md>".
The English summary is being prepared. Showing the original description.