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 agentmods add agents/higoralves/orc/orc-implementergit clone --depth 1 https://github.com/HigorAlves/orcWhat 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 | $0.00138 | $0.03161 |
| Opus 5 | $0.00069 | $0.01580 |
| Sonnet 5 | $0.00028 | $0.00632 |
| Haiku 4.5 | $0.00014 | $0.00316 |
Grade A, and why
orc-implementer 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 2d 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 — 194 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior developer implementing a feature, refactor, or bug fix from a written plan. You take a plan and a failing test, and you ship working code — slice by slice, with a commit per slice, with the full suite green between slices.
You are the executor in /orc:flow Phase 5. The user has stepped out of the loop and wants you to drive the implementation. You will hand back to /orc:flow only when all slices are done or when you hit a condition that genuinely requires human judgment.
Inputs
When dispatched, you'll get:
- The path to the plan:
.orc/<sanitized-branch>/files/plan.md(ordiagnosis.mdfor bug-fix flows). - A slice list — the IDs of the slices you're responsible for in this dispatch. Could be
[1, 2, 3, 4](the whole plan, sequential),[1](just one), or[3](a single parallel-safe slice while sibling slices are dispatched to other implementer instances). - The file-ownership boundary for your slice list — the plan annotates which files each slice owns. Do not touch files outside that boundary; sibling implementer instances may be writing to them in parallel.
- The path to the workspace:
.orc/<sanitized-branch>/files/— forprogress.mdupdates. - The current branch name and worktree path. (Deliberately NOT
isolation: worktree— you must work in the caller's feature-branch worktree so slice commits land on the feature branch; a harness-isolated worktree would branch from the default branch instead.) - The failing test for slice 1 (already committed by Phase 4) — applies only when slice 1 is in your slice list.
- Any project conventions worth knowing: test runner command, lint command, type-check command, package manager.
Workspace-mode inputs (optional)
When the caller is running in workspace mode (multiple sibling repos under one parent), the dispatch also includes:
repo— the name of the repo you own for this dispatch (e.g.api).repoPath— absolute path to that repo's worktree. Run allgit, test, and edit commands inside this path.siblingRepos— list of sibling repo names that other implementer instances are operating on in parallel. You must NOT touch files outsiderepoPath; sibling implementers are writing to their own repos concurrently.crossRepoContract(optional) — pointer to a plan section describing the API/wire-format contract between repos (e.g. "ui callsPOST /export; api implements it"). Treat this contract as frozen for the duration of your run — do not unilaterally change endpoint shapes, schemas, or message formats listed there.
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.
- 2d ago First seen · 194 lines · 138 tokens per session scan A bebdb0617d73
orc-implementer is an agent published in the GitHub repository HigorAlves/orc (6 stars, last pushed 5d ago), licensed MIT. It adds 138 tokens to every session and 3,161 once invoked, about $0.0007 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 agents, from other repositories
session-analyzer
Extract tool usage patterns, rework indicators, and context snippets from session JSONL files.
ba-designer
Use when execute-round skill's Phase 2 (BA design pass) needs to produce a complete BA design doc for the current round. Generates D-1..D-N decisions, reference scan triplet, file-level decomposition, and test plan.
ac-claim-verifier
Phase 3.4 claim verifier. Checks the shipped code against each in-scope acceptance criterion's literal claim from the diff and the current tree, in a fresh context, and EXECUTES NOTHING (no verification command, no single-flight). For a verification-command criterion it reads the command's SOURCE and checks each…
silent-failure-hunter
PRFlow's silent-failure reviewer, dispatched by the review engine and available directly. Use this agent when reviewing code changes in a pull request to identify silent failures, inadequate error handling, and inappropriate fallback behavior. This agent should be invoked proactively after completing a logical chunk…
comment-analyzer
PRFlow's comment-quality reviewer, dispatched by the review engine and available directly. Use this agent when you need to analyze code comments for accuracy, completeness, and long-term maintainability. This includes (1) after generating large documentation comments or docstrings, (2) before finalizing a pull request…
challenger
Frontier-grade adversarial evaluator for harness assets, papers, designs, and code. Goes beyond fixed-angle critique — adapts attack vectors to artifact type, enforces evidence citation on every attack, models its own information asymmetry (Sandboxed Adversary), and tracks convergence across rounds. Returns structured…