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 antoinedc/MantaUI --skill manta-handle-reviewer-returngit clone --depth 1 https://github.com/antoinedc/MantaUIWrote 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/antoinedc/mantaui/manta-handle-reviewer-return)<a href="https://agentmods.dev/skills/antoinedc/mantaui/manta-handle-reviewer-return"><img src="https://agentmods.dev/badge/skills/antoinedc/mantaui/manta-handle-reviewer-return/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/antoinedc/mantaui/manta-handle-reviewer-return"><img src="https://agentmods.dev/badge/skills/antoinedc/mantaui/manta-handle-reviewer-return.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Tool Misuse · line 103 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
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 | $0.00069 | $0.01510 |
| Opus 5 | $0.00034 | $0.00755 |
| Sonnet 5 | $0.00014 | $0.00302 |
| Haiku 4.5 | $0.00007 | $0.00151 |
Grade A, and why
manta-handle-reviewer-return 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 12d 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 — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
manta-handle-reviewer-return
If the reviewer reassigns an issue back to you (status todo,
your PR still open in the comment chain), read loop_history FIRST,
before doing anything else. This is how the loop learns — without it,
each return starts from zero and repeats the same mistake.
export MULTICA_WORKSPACE_ID=264c89bb-4659-4570-af7b-5f8daaf87985
multica issue metadata get <ISSUE-KEY> --key loop_history --output json
Read the history. For each prior attempt, note:
- What you tried (
approachfield) — do NOT repeat this approach. - What was weak (
self_checkfield — the criterion that scored lowest). - The reviewer's actual Block/Question from each cycle — these are the ground-truth failures, more reliable than your memory of fixing them.
If the SAME Block reappeared after you "fixed" it, that's your signal: your fix didn't address the root cause. See "Recurring / phantom Block" below for diagnosis. Do not start fixing until you've read this.
Then read the reviewer's most recent Multica comment to identify which kind of return.
Two cases
"Review found N Block-severity findings"
For each Block, check the scope suffix:
-
Block (fix-here)→ fix the code per the Block list in the PR review, push to the same branch, then reassign back to the reviewer. Do not open a new PR. -
Block (followup-issue)→ do NOT try to fix it in this PR. The reviewer's PR comment includes a one-paragraph description meant for a new issue body. File the new issue viamultica issue create --title "<reviewer's title>" --description-stdin <<< "<reviewer's paragraph>" --status todo --priority medium(unassigned by default — the human triages). Post a PR commentAddressed via BET-XX (separate issue). Resuming this PR., then reassign this issue back to the reviewer for the next review pass. -
Don't conflate the two. Treating a
Block (followup-issue)as if it werefix-hereis the costliest failure mode — unbounded scope creep.
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.
- 12d ago First seen · 145 lines · 69 tokens per session scan A 2558c8f3e987
manta-handle-reviewer-return is a skill published in the GitHub repository antoinedc/MantaUI (12 stars, last pushed 6d ago), licensed MIT. It adds 69 tokens to every session and 1,510 once invoked, about $0.0003 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
code-review
Reviews code changes, pull requests, and diffs for correctness, security, performance, and style. Use when the user submits a PR for review, asks to review a diff or code snippet, or requests a quality check on recent changes.
gh-find-prs
Survey open Codewhale PRs and triage each for mergeability and disposition against the real landing branch.
plannotator
Reference for using the Plannotator CLI: plan review, code review, annotating files, URLs, folders, and running local apps, annotating the last assistant message, browsing archived plan decisions, and exporting or sharing Guided Reviews. Invoke when asked to use Plannotator for anything not covered by a more specific…
find-simplifications
Use for a periodic repo-wide sweep of qwen-code for accumulated excess surface — dead components and files, orphaned locale keys, exports nothing consumes, added-then-removed scaffolding — filing candidates on a tracking issue and landing only what a maintainer has said yes to. Repo-wide and evidence-first; every…
review-work
Post-implementation gate review: run manual QA on the real surface yourself, then launch ONE gate reviewer (never a panel) to audit goal, constraints, code quality, security, missed context, and QA evidence. Use before a PR handoff or when the user explicitly asks to review completed work.
agtx-review
Self-review completed work. Check for correctness, edge cases, and code quality. Write review to .agtx/review.md and stop.