Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add jpbaking/goal-ledger --skill goal-ledger-resumegit clone --depth 1 https://github.com/jpbaking/goal-ledgerWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jpbaking/goal-ledger/goal-ledger-resume)<a href="https://agentmods.dev/skills/jpbaking/goal-ledger/goal-ledger-resume"><img src="https://agentmods.dev/badge/skills/jpbaking/goal-ledger/goal-ledger-resume/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jpbaking/goal-ledger/goal-ledger-resume"><img src="https://agentmods.dev/badge/skills/jpbaking/goal-ledger/goal-ledger-resume.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00113 | $0.01322 |
| Opus 5 | $0.00056 | $0.00661 |
| Sonnet 5 | $0.00023 | $0.00264 |
| Haiku 4.5 | $0.00011 | $0.00132 |
Grade A, and why
goal-ledger-resume scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Goal Ledger — resume and recover
Treat the ledger and Git as evidence left by another agent — possibly a weaker one, possibly mid-sentence. Establish what is verified, repair bookkeeping without discarding work, and continue from the first safe action.
1. Establish your role
- If you were handed a specific phase to run, use the
goal-ledger-executeskill instead. - If you are an ordinary helper subagent working from a delegated prompt, ignore the ledger and this skill and continue only the task you were given. An unfinished ledger is not a blocker for that task and is not worth reporting unless the prompt asks for ledger work.
- Otherwise you are resuming as the planner: you own the goal, the plan,
Goal status, Git, and Gate D.
Read sections 1–4 of the sibling goal-ledger skill. If .goal-ledger/GOAL.md is missing, say there is no ledger to resume. If status is completed or abandoned, report that and stop.
2. Decide whether this request resumes the goal
- The user asks to resume, or the request matches the recorded Outcome: resume without another intent question.
- The ledger is
blocked-on-humanand the user answers its blockers: record the answers into the affected phase's## Contextso a re-delegated executor can see them, return affected items to[todo], and continue. - Status
drafting: return to Gate A. Statusapproved: if strategy, branch, baseline, or approval commit is missing or inconsistent, return to Gate B; otherwise return to Gate C. - Status
in-review: Gate D was started and not finished. Re-run it from the top rather than trusting a partial## Review. - Status
awaiting-acceptance: report the result and ask for acceptance; do not resume execution. - An unrelated request: report the unfinished goal and ask whether to resume
it or use the
goal-ledger-abandonskill. Never overwrite it silently.
3. Reconcile Git
Skip Git steps when Repository: no.
- Verify the current branch equals
Work branch. If not, do not switch with a dirty tree. Report both branches and ask before switching. - Verify the immutable baseline exists and is an ancestor of
HEAD. If not, stop with the exact inconsistency. - Inspect
git log <baseline>..HEAD, including commit trailers. Classify matching Goal-ID commits, foreign commits, merge commits, and phase begin/close pairs. Record foreign history in Handoff; it may prohibit squashing but does not justify deleting work. - Compare locally known starting/work upstream refs with their recorded Gate B SHAs. Record advances, rewinds, divergence, or goal-commit publication in Handoff. On
current-branch, any upstream movement prohibits automated squash. - Inspect the worktree. A phase-begin commit followed by a clean tree means phase work did not start. A dirty tree means work may have stopped mid-phase; inspect the diff and do not discard it.
- A completed phase without a matching close commit is unverified until its
## Verifypasses. A close commit with stale ledger status may be reconciled after its check passes.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 63 lines · 113 tokens per session scan A 83b8135aba43
goal-ledger-resume is a skill published in the GitHub repository jpbaking/goal-ledger (1 stars, last pushed 1mo ago), licensed 0BSD. It adds 113 tokens to every session and 1,322 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
mulmoterminal-keys
Bind keyboard shortcuts and fix keyboard/clipboard behaviour in MulmoTerminal. Writes keymap, which Settings cannot set at all — its Keyboard shortcuts section is read-only, listing every action bound or not plus a send row. Explains copyOnSelect, questionPaneEnabled and terminalSubmit, which have their own Settings…
template-engine
Load, validate, and populate templates consistently across frameworks.
checkpoint
Create, list, or recover mid-workflow checkpoints so interrupted work resumes from a known-good position.
update
Update AIWG through its install-aware updater and re-deploy all installed frameworks from the registry.
run
Execute a named script defined in .aiwg/aiwg.config or list all available scripts.
runtime-info
Display a comprehensive runtime environment report covering detected provider, available tools, AIWG version, and installed frameworks.