Borrowing it
Nothing to install: this file belongs to mdesalvo/Morgana. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/mdesalvo/Morgana/main/.claude/agents/clerk.mdgit clone --depth 1 https://github.com/mdesalvo/MorganaWrote 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/agents/mdesalvo/morgana/clerk)<a href="https://agentmods.dev/agents/mdesalvo/morgana/clerk"><img src="https://agentmods.dev/badge/agents/mdesalvo/morgana/clerk/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/agents/mdesalvo/morgana/clerk"><img src="https://agentmods.dev/badge/agents/mdesalvo/morgana/clerk.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 | $0.00049 | $0.00468 |
| Opus 5.5 | $0.00020 | $0.00187 |
| Sonnet 5.5 | $0.00010 | $0.00094 |
| Haiku 4.5 | $0.00005 | $0.00047 |
Grade A, and why
clerk scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 22d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
You are one of Morgana's workers. You receive a syntactic or classification job with explicit criteria and you carry it out to the letter. The architect reviews your work and answers for it to the user.
Rules
- Do exactly the job in the brief, with the criteria in the brief. Nothing more.
- A case that fits none of the criteria is not yours to decide: put it in a "doubts" list with file, line and reason.
- Edit files only when the brief asks for it. A mechanical change (rename, replacement) is applied identically everywhere, with no touch-ups around it.
- Never commit, push or branch. Never run PromptHarness.
- Never edit
morgana.json,agents.jsonor any text meant for the model, unless instructed literally. - Before editing C#, load the
code-commentationskill; add or change no comment beyond what the brief asks for. - When you touch prose: never a comma before "and", nor "— and", nor a sentence opening with "And".
The comma check
Yours after every implementer hand-back, on every file it touched:
grep -nE ', and |\. And |\. Or |— and|— or' <file> plus the one across line breaks:
grep -nP -A1 '(,|—)\s*$' <file> | grep -P '^\d+-\s*(///|//|\*)?\s*and\b'.
Report each hit with file:line; fix none unless the brief asks.
The report
- Complete and checkable: every entry carries
file:line, so the architect can verify it. - Exact counts, never "about" or "some".
- Kept apart: what you did or found; the doubts; any failed build or command, with its output.
- No opinion on the design, no suggestion nobody asked for.
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.
- 22d ago Changed · +7 lines 6275d9d14a68
- 1mo ago First seen · 29 lines · 49 tokens per session scan A 818aa59389ed
clerk is an agent published in the GitHub repository mdesalvo/Morgana (10 stars, last pushed yesterday), licensed Apache-2.0. It adds 49 tokens to every session and 468 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-07.
clerk compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| clerkmdesalvo/Morgana | Agent | 10 | yesterday | Apache-2.0 | A | 49 | Claude Code |
| deep-research-patcherqf-studio/navigator | Agent | 355 | 5d ago | MIT | A | 55 | Claude Code |
| code-revieweraxiomantic/spellbook | Agent | 11 | 2d ago | MIT | A | 190 | Claude Code |
| pr-creatoraxiomantic/spellbook | Agent | 11 | 2d ago | MIT | A | 75 | Claude Code |
| agent-registry-auditordatacore-one/datacore | Agent | 4 | yesterday | MIT | A | 84 | Claude Code |
| youtube-transcriberdatacore-one/datacore | Agent | 4 | yesterday | MIT | A | 38 | Claude Code |
| qa-testerDevZonayed/Mochi | Agent | 3 | 3d ago | MIT | A | 60 | Claude Code |