Getting it into your agent
There is no command for this one: it runs only inside a plugin, and the catalogue could not identify which plugin ships it. The source is linked below.
Wrote 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-pr-workflow)<a href="https://agentmods.dev/skills/antoinedc/mantaui/manta-pr-workflow"><img src="https://agentmods.dev/badge/skills/antoinedc/mantaui/manta-pr-workflow/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-pr-workflow"><img src="https://agentmods.dev/badge/skills/antoinedc/mantaui/manta-pr-workflow.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 | $0.00071 | $0.04774 |
| Opus 5 | $0.00036 | $0.02387 |
| Sonnet 5 | $0.00014 | $0.00955 |
| Haiku 4.5 | $0.00007 | $0.00477 |
Grade A, and why
manta-pr-workflow 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 10d 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 — 356 lines — stays where its author put it; the contents beside it link to each section on GitHub.
manta-pr-workflow
The standard implementer workflow for the MANTA agents.
Steps
-
multica repo checkout [email protected]:antoinedc/MantaUI.gitinside your workdir. Never edit files outside the workdir. (A human-owned checkout at/home/dev/projects/better-uibypasses isolation.) -
Sync
maintoorigin/mainBEFORE you branch (mandatory — stale-base antibody). A reused/cached workdir can hold amainthat is days behindorigin/main; branching off it makes your diff appear to delete files that were merged toorigin/mainafter your stale checkout, producing phantom "you deleted X" review Blocks that no fix can resolve. Always:git fetch origin git checkout main git reset --hard origin/main # workdir is disposable; make main == origin/main git checkout -b multica/BET-N-<slug> # Record the base you branched from, for the PR body (step 11): git rev-parse --short origin/mainThe recorded
origin/mainshort-sha is your base SHA — put it in the PR body so the reviewer can confirm freshness in one glance. -
Read the issue. If it touches IPC, renderer, main process, preload, or server routes, re-read the matching architecture docs in the repo.
-
Run the anti-spaghetti contract (below) before writing any new code. Search for an existing implementation first; the default move is to extend/reuse, not to add a parallel copy.
-
Implement. Add or update tests under
tests/or__tests__/.
5b. Commit incrementally while you implement (BET-1445 — worktree-hijack
insurance). A sibling session's multica repo checkout can resolve to
YOUR worktree mid-run and hard-reset it (git reset --hard origin/main +
branch switch), destroying uncommitted edits (killed the BET-1439 rerun,
~45 min). Checkouts are git worktrees of a shared bare clone, so
committed work on your task branch survives a hijack — only the
uncommitted delta is lost. So: commit to your task branch after every
coherent unit and never leave more than a few minutes of work uncommitted
(then push per step 8). And if your worktree suddenly looks reset/switched
(unexpected origin/main HEAD, foreign branch, edits gone): run
git branch --show-current, then recover — git checkout multica/BET-N-<slug> in place (or git worktree add <dir> multica/BET-N-<slug> if checkout is disturbed) — verify with git log
that your commits are intact, redo only the lost delta, and CONTINUE. Do
not assume the work is lost and restart from zero.
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.
- 10d ago First seen · 356 lines · 71 tokens per session scan A ae5ab3966a19
manta-pr-workflow is a skill published in the GitHub repository antoinedc/MantaUI (12 stars, last pushed 4d ago), licensed MIT. It adds 71 tokens to every session and 4,774 once invoked, about $0.0004 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
tessera-cli
Operate Tessera-managed Projects, Worktrees, and Sessions through the injected version-matched control CLI. Use when an agent running inside Tessera needs to inspect its caller context, create an isolated Worktree, launch or observe a Session, send a follow-up at a known input boundary, or stop a live runtime.
repo-hygiene
Use when the scheduled repo-hygiene workflow runs from GitHub Actions (or an operator dry-run) to scan the repository for small, certain docs/test/code hygiene issues and fix them as one batched branch.
create-issue
Draft and submit a GitHub issue from a user idea or bug description, with bilingual body and correct labels.
project-session-manager
Worktree-first dev environment manager for issues, PRs, and features with optional tmux sessions.
cw-land
Use when turning verified Codewhale work into commits, branches, or a merge: choosing direct-main vs. worktree vs. integration branch, preserving contributor credit, and honoring the gate artifact before merging.
cw-gates
Use before claiming any Codewhale change is done, green, or ready to land: the focused-to-broad verification ladder, the budget checks CI enforces, and the rules for what counts as a passing test.