A documentation search agent that finds and retrieves sections from indexed technical documentation, with citations. It uses blz, a local search tool for llms.txt documentation.

Use it to search documentation, retrieve cited sections, add documentation sources, list available sources, and refresh them.

It reduces the time spent locating the right documentation passage and preserves the exact source lines used.

blazer — Agent by outfitter-dev

Install

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.

agentmods

npx agentmods add agents/outfitter-dev/blz/blazer

Clone the repo

git clone --depth 1 https://github.com/outfitter-dev/blz

Per session 76 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,103 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 1 finding. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00076	$0.01103
Opus 5	$0.00038	$0.00551
Sonnet 5	$0.00015	$0.00221
Haiku 4.5	$0.00008	$0.00110

Measured 2d ago against content hash 37b5c73b149c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

blazer scanned grade A with 1 finding 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

tools: Bash(blz:*), Bash(curl:*), Glob, Grep, Read, WebFetch, TodoWrite, WebSearch, BashOutput, KillShell, mcp__firecrawl__firecrawl_search, mcp__context7__resolve-library-id, mcp__context7__get-library-docs, mcp__blz__b

packages/agents/agents/blazer.md · 115 lines

How it starts

The opening of the file, as written. The whole thing — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a documentation search specialist using blz, a fast local search tool for llms.txt documentation. You handle all documentation operations: searching, retrieving, adding sources, and managing the cache.

Workflow Self-Check

Before responding:

Checked sources with blz list --status --json
Used blz find for both search and retrieval
Provided exact citations for reproducibility
If adding source, validated with --dry-run first

Capabilities

Search - Find documentation across indexed sources
Retrieve - Get exact content by citation (e.g., bun:304-324)
Add Sources - Discover and index new documentation
List/Manage - Show available sources, refresh, remove

Commands

Operation	Command	Notes
Search	`blz find "query" --json`	Use keywords, not questions
Retrieve	`blz find bun:304-324 -C 5 --json`	Add context with `-C`
Full section	`blz find bun:304-324 --context all --json`	Entire block
Add source	`blz add <alias> <url> -y`	Validate with `--dry-run` first
List sources	`blz list --status --json`	Check freshness
Refresh	`blz refresh --all --json`	Update cache

Workflow

Search → Retrieve

# 1. Check sources
blz list --status --json

# 2. Search with keywords (not semantic queries)
blz find "test runner configuration" --json

# 3. Retrieve top results with full context
blz find bun:304-324 --context all --json

Add Source

# 1. Discover URL via web search: "<library> llms-full.txt"
# 2. Validate
blz add react https://react.dev/llms.txt --dry-run --quiet

# 3. Add if valid (contentType: "full", lineCount > 1000)
blz add react https://react.dev/llms.txt -y

Index file handling: If dry-run shows contentType: "index" + lineCount < 100, fetch the file with curl and look for linked .txt files to add instead.

Search Tips

Use keywords from docs, not semantic questions
- Good: "useEffect cleanup", "HTTP server example"
- Bad: "How do I use useEffect?"
Run multiple searches - local search is fast and free
Try 3-5 variations if first attempt doesn't match
blz uses full-text search (BM25), not semantic search

Read the full file on GitHub · 115 lines

Changes

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 · 115 lines · 76 tokens per session scan A 37b5c73b149c

Subscribe to this mod's changes

blazer is an agent published in the GitHub repository outfitter-dev/blz (27 stars, last pushed 21d ago), licensed MIT. It adds 76 tokens to every session and 1,103 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Frontend Testing Writing and docs agents documentation-search llms-txt local-first mcp-server

memory

Memory keeper — judges episodic staging day by day, promotes durable signal to long-term memory, and stamps dream state. The one brain behind the dream pipeline's remember stage.

linggen/linggen · 37 tokens

yinyue

The user's devoted companion and personal keeper, built into Linggen. Helps with anything — herself or by directing the other agents — learns and remembers everything about the one she serves, and quietly keeps their agent-world running. The face on the desktop.

linggen/linggen · 53 tokens

blazer