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 commands/outfitter-dev/blz/discovergit clone --depth 1 https://github.com/outfitter-dev/blzWhat 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 | $0.00000 | $0.00788 |
| Opus 5 | $0.00000 | $0.00394 |
| Sonnet 5 | $0.00000 | $0.00158 |
| Haiku 4.5 | $0.00000 | $0.00079 |
Grade A, and why
discover 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.
What it actually says
Registry: Discover Full Documentation Sources
Instructions: $ARGUMENTS
- Invoke the
@agent-registry-handlersubagent to discover full documentation sources from index files - Provide the subagent with:
- The source ID or URL to analyze (from $ARGUMENTS)
- Any specific instructions about what to look for or focus on
- Context about recent changes or known issues with the source
Instructions to provide subagent:
-
Resource: Use https://llmstxthub.com/ as a reference to discover additional llms.txt sources or verify known sources
- Search the hub for the project name to find official llms.txt URLs
- Check if the hub lists any additional documentation files beyond the main index
- Cross-reference discovered URLs with the hub's listings
-
If given a source ID (e.g., "redis"):
- Run
./registry/scripts/discover-links.sh <source-id>to automate initial discovery - Analyze the output and identify all discovered .txt links
- Test each link with
blz add --dry-runto classify content type - Cross-check with https://llmstxthub.com/ for any additional sources
- Report findings with recommendations for which sources to add
- Run
-
If given a URL:
- First run
blz add temp-<name> <url> --dry-run --quietto analyze content - If contentType is "index" (< 100 lines), proceed with discovery workflow
- Extract all .txt references from the index file
- Resolve relative URLs to absolute URLs
- Cross-check with https://llmstxthub.com/ to find additional sources not listed in the index
- Test each discovered URL with dry-run analysis
- Report findings with actionable recommendations
- First run
-
For each good candidate (contentType: "full", lines > 1000):
- Provide a recommended
blz registry create-sourcecommand - Include suggested metadata (description, category, tags, aliases)
- Note any special considerations (redirects, format variations, etc.)
- Provide a recommended
Follow-up steps after discovery:
-
Present a summary report showing:
- Total sources discovered
- Good candidates (recommend adding)
- Mixed content (needs review)
- Indexes (skip)
- Any issues or edge cases encountered
-
For each good candidate, show the complete command:
blz registry create-source <name> \ --url <url> \ --description "<description>" \ --category <category> \ --tags <tags> \ --npm <packages> \ --github <repos> \ --yes -
If the original source is an index that should be kept:
- Note that it needs "(Index)" suffix in name
- Note that it needs "index" tag added
- Provide edit instructions or offer to update it
-
After user confirms which sources to add:
- Execute the
blz registry create-sourcecommands - Update the original index entry (if keeping it)
- Run
./registry/scripts/build.shto rebuild registry - Verify all new sources are in
registry.json
- Execute the
-
Ask the user:
- "Would you like me to proceed with adding these sources to the registry?"
- "Should I keep the original index entry or remove it?"
- "Are there any other sources you'd like me to analyze?"
Example usage:
/registry discover redis- Discover from existing registry source/registry discover https://anthropic.com/llms.txt- Discover from URL/registry discover check llmstxthub.com for new sources- Browse the hub for new projects to add
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 · 77 lines · 0 tokens per session scan A 48958eeedc9e
discover is a command published in the GitHub repository outfitter-dev/blz (27 stars, last pushed 21d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 788 tokens. 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 2026-08-30.
Other commands, from other repositories
CAA-add
Add a valuable fix/insight from this session to the Claude Archive (CAA) — give, score honestly, sync.
CAA-search
Search the Claude Archive (CAA) for scored solutions BEFORE solving — take the best hit, verify, apply.
CAA-add-latest
Instantly archive the LATEST hard-won fix from this session to the Claude Archive (CAA) — zero questions.
template
Manage issue templates for streamlined issue creation.
cross-validate
Take a specific scientific claim and confirm or refute it across 3+ independent databases, then report concordance. Use before publishing, citing, or acting on a fact when you want to know how strongly it's supported. Forces multi-source verification that the agent doesn't naturally enforce.
translate-id
Resolve an identifier across all relevant namespaces (HGNC symbol, Ensembl, UniProt, NCBI Gene ID, RefSeq, MGI, OMIM, ChEMBL, PubChem, etc.). Detects the input namespace automatically, picks the right resolver tool, and returns a complete cross-reference table. Use when you have an ID in one namespace and need it in…