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.
git clone --depth 1 https://github.com/nguyenthienthanh/aura-frogWrote 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/commands/nguyenthienthanh/aura-frog/heal)<a href="https://agentmods.dev/commands/nguyenthienthanh/aura-frog/heal"><img src="https://agentmods.dev/badge/commands/nguyenthienthanh/aura-frog/heal.svg" alt="Measured on agentmods" 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 | $0.00000 | $0.01071 |
| Opus 5 | $0.00000 | $0.00535 |
| Sonnet 5 | $0.00000 | $0.00214 |
| Haiku 4.5 | $0.00000 | $0.00107 |
Grade A, and why
heal 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 7d 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/aura-frog:heal
Self-healing orchestration. Diagnose F2/F3 failures, view proposals, accept/decline, disable.
Usage
/aura-frog:heal diagnose <TASK_ID> # run self-healing-orchestrator on the failed task; emit proposal
/aura-frog:heal status # session counter, budget, last proposal, disabled state
/aura-frog:heal accept <HEAL-ID> # apply a proposal as a new T4 task with its own approval flow
/aura-frog:heal decline <HEAL-ID> [reason] # reject; counts toward session cap
/aura-frog:heal disable # disable for this session (.claude/logs/.self-heal-disabled flag)
/aura-frog:heal enable # re-enable for this session
Protocol — diagnose
- Refuse if
AF_SELF_HEAL_DISABLED=trueor.claude/logs/.self-heal-disabledexists - Refuse if session cap reached (5 proposals total)
- Read failure-classifier output for the task — refuse if class ∉ {F2, F3}
- Refuse if task already has an accepted heal (max 1/task)
- Check
replan_budget_remainingon the task; refuse if 0 - Invoke
self-healing-orchestratorskill — it produces a proposal artifact at.claude/plans/proposals/HEAL-<TASK_ID>-<NNN>.yaml - If confidence < 0.7: surface raw findings only ("self-heal couldn't reach a confident proposal — escalating findings"); do NOT create a proposal
- If confidence ≥ 0.7: surface diff + reasoning + risks; prompt user to
/aura-frog:heal acceptor/aura-frog:heal decline - Append history.jsonl event:
event: self_heal_proposed
Protocol — accept <HEAL-ID>
- Refuse if proposal not found at
.claude/plans/proposals/HEAL-<HEAL-ID>.yaml - Refuse if already applied (idempotency check)
- Save checkpoint on the affected node (per
rules/workflow/checkpoint-discipline.md) - Apply the patch as a NEW T4 task (mint TASK-NNNNN), NOT a modification of the failing task
- The new T4 task goes through the standard approval flow (Phase 1 design optional, Phase 2/3 mandatory)
- Decrement
replan_budget_remainingon the parent T3 (counts as a replan) - Increment session counter
- Append history.jsonl:
event: self_heal_accepted, link to original proposal + new task ID
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.
- 7d ago First seen · 85 lines · 0 tokens per session scan A 100c5b8c37e2
heal is a command published in the GitHub repository nguyenthienthanh/aura-frog (24 stars, last pushed 3d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,071 tokens. 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 2026-08-30.
Other commands, from other repositories
research_codebase
Document codebase as-is with thoughts directory for historical context.
improve
/improve Scan and fix all improvement categories /improve arch Architectural improvements only /improve duplication Duplicate code / abstraction consolidation /improve quality Code quality (GuardLoop patterns GL-01 through GL-10) /improve security Security posture improvements /improve --dry-run Show what would be…
cleanup
Detect and remove orphaned code, unused components, dead routes, and stale database artifacts.
poly-check
Lint and check formatting with poly — apply no fixes; summarize findings and drift.
logic-health
Sweep a whole codebase or directory for logic correctness — use before a release or to identify risk hotspots.
validate
Run SDLC compliance check against the current project. Validates build system, code quality, testing, CI/CD, security, documentation, VCS, and release configurations.