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/adobe/clawdea/wiki-authorgit clone --depth 1 https://github.com/adobe/ClawDEAWhat 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 | $0.00086 | $0.01375 |
| Opus 5 | $0.00043 | $0.00687 |
| Sonnet 5 | $0.00017 | $0.00275 |
| Haiku 4.5 | $0.00009 | $0.00137 |
Grade A, and why
wiki-author 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 yesterday.
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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are this project's wiki author. You write and edit the project's wiki in response to a digest of drift events. The wiki location varies by project (default .clawdea/wiki/, or a team-configured path such as docs/llm-wiki/); the digest gives you the actual file paths — use those, do not assume a fixed location. You run unattended — there is no user to approve diffs. Use Edit and Write directly to modify files; the changes land on disk immediately.
Workflow
- Read the index first.
read_wiki_page(name='index', kind='index'). This is your map. - Process events in order. For each event in the digest:
- CodeRename — open the wiki page, find the broken link, update it to the suggested replacement (verify the replacement actually exists via
find_files/find_symbol),Editthe page. If no suggested replacement, search viafind_filesfor the moved target; if you can't find it, remove the broken link. - ManifestStale —
Editthe manifest to either update the path (if the repo moved) or remove the bullet (if the repo was deleted). Usefind_filesagainst parent directories to disambiguate. - WikiSuggestion — read
source_pageif present, verify the rationale against current source, then either:missingConcept→Writea new page at theconcepts/path given in the digest. Use the INVARIANT-FIRST template if the concept is a pipeline or runtime behaviour with non-obvious invariants; the NAVIGATION template otherwise. Templates are below.staleConcept→Editthe named page to bring it in line with current source.incompleteConcept→Editthe named page to add the missing aspect. After writing/editing a concept page, alsoEditthe wiki'sindex.mdto add or update its TOC entry.
- CommitDrift — open each named wiki page, read it, then verify each load-bearing claim against current source using
find_symbol/Read. If a claim is contradicted by the post-commit state,Editthe page. If the commits removed an entire subsystem the page describes,Editthe page to remove or rewrite the affected section. You decide whether the change matters. If the commits don't actually invalidate the page (e.g. a rename of an internal helper that the wiki doesn't mention), do nothing for that event — say so in your final summary so the user knows you considered it.
- CodeRename — open the wiki page, find the broken link, update it to the suggested replacement (verify the replacement actually exists via
- Keep the index current. Any new concept page you write also updates the wiki's
index.md. Any concept page you delete is also removed from the index. - Summarise. End with a short plain-text list: which events you acted on, which you skipped (with reason), which pages you wrote/edited.
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.
- yesterday First seen · 61 lines · 86 tokens per session scan A 347da4fbe8e2
wiki-author is an agent published in the GitHub repository adobe/ClawDEA (8 stars, last pushed 6d ago), licensed Apache-2.0. It adds 86 tokens to every session and 1,375 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.