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 instructions/recondodev/recondo/claude-mdgit clone --depth 1 https://github.com/recondodev/recondoWrote 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/instructions/recondodev/recondo/claude-md)<a href="https://agentmods.dev/instructions/recondodev/recondo/claude-md"><img src="https://agentmods.dev/badge/instructions/recondodev/recondo/claude-md.svg" alt="Measured on agentmods" 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 | $0.03795 | $0.03795 |
| Opus 5 | $0.01898 | $0.01898 |
| Sonnet 5 | $0.00759 | $0.00759 |
| Haiku 4.5 | $0.00380 | $0.00380 |
Grade A, and why
recondo CLAUDE.md 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 4d 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 — 315 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Recondo
AI Governance Gateway. Every agent-to-LLM call flows through it for compliance auditing (SOC 2, ISO 42001), usage intelligence, and centralized reporting across all platforms.
Build & Run
just setup # one-time: installs cargo-nextest
just build # fmt + clippy + build
just run # starts gateway on :8443
just test # runs all 678 tests via nextest
just ci # fmt + clippy + test (what CI runs)
Or without just:
cd gateway && cargo build
cd gateway && cargo run
cd gateway && cargo nextest run --features test-support
cd gateway && cargo clippy --features test-support --tests -- -D warnings
Test the gateway
Automated tests (678 tests, ~5 seconds)
just test
Run a specific test suite:
cd gateway && cargo nextest run --test tls_mitm_tests # TLS MITM pipeline (8 tests)
cd gateway && cargo nextest run --test tcp_listener_tests # TCP listener (6 tests)
cd gateway && cargo nextest run --test wal_tests # WAL module (12 tests)
cd gateway && cargo nextest run --test session_tests # Session management (16 tests)
cd gateway && cargo nextest run --test capture_integration_tests # Capture pipeline (8 tests)
Test suite map
| Test file | Tests | What it covers |
|---|---|---|
week4_tests |
57 | Schema v2 migration, Gemini provider/endpoint, fixture validation, compliance docs |
week4_phase2_tests |
53 | Gemini response parser, mock LLM servers, struct v2 wiring |
schema_expansion_tests |
51 | Schema field coverage, DB roundtrips |
proxy_tests |
37 | CONNECT request parsing, host classification, server config |
intercept_tests |
37 | Request interception: path matching, method filtering, Gemini endpoints |
cli_tests |
28 | CLI commands: sessions, session, turn, search, stats, verify |
websocket_tests |
26 | WebSocket frame parsing, encoding, masking |
code_review_fixes_tests |
24 | Fixes from prior code reviews |
db_tests |
18 | SQLite schema, insert/query, v2 struct round-trips |
session_tests |
16 | Session boundaries, prompt hash, time gaps, sequence |
schema_full_tests |
15 | Full schema validation with all fields |
websocket_integration_tests |
14 | WebSocket relay, frame capture, DB integration |
intent_agent_tests |
14 | Initial intent extraction, agent framework detection |
anthropic_response_tests |
14 | SSE event parsing: text, thinking, tool_use, usage |
week2_negative_tests |
13 | Week 2 negative/edge case tests |
stream_tests |
13 | SSE accumulator: chunking, message_stop, truncation |
wal_tests |
12 | WAL: append, flush, mark_flushed, persistence, fail modes |
gap_fix_tests |
12 | Gap fixes from compliance analysis |
anthropic_request_tests |
12 | Request body parsing: system prompt, messages, tools |
migration_tests |
11 | Additive migration framework, idempotency |
forward_compat_tests |
11 | OD-007: raw_extra, parser_version, parse_errors |
tls_tests |
10 | CA generation, leaf cert generation, extra CA cert loading |
store_tests |
9 | Object store: gzip, content-addressable, dedup |
tls_mitm_tests |
8 | End-to-end TLS MITM: CONNECT → TLS handshake → capture to disk |
messages_delta_tests |
8 | Messages delta compression, DB roundtrip |
capture_metadata_tests |
8 | Capture record metadata fields |
capture_integration_tests |
8 | Full capture pipeline: req → hash → store → DB |
provider_tests |
7 | Provider detection: Anthropic, OpenAI, Gemini |
negative_tests |
7 | Negative/boundary tests |
hash_tests |
7 | SHA-256 computation |
tcp_listener_tests |
6 | TCP listener: bind, accept, CONNECT 200, concurrent connections |
pipeline_tests |
5 | Capture pipeline orchestration |
schema_tests |
3 | Schema type definitions |
gemini_integration_tests |
3 | Gemini process_capture → DB round-trip (text, tool calls, raw_extra) |
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.
- 4d ago First seen · 315 lines · 3,795 tokens per session scan A b30a84d00b54
recondo CLAUDE.md is an instructions file published in the GitHub repository recondodev/recondo (5 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 3,795 tokens to every session, about $0.0190 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 instructions, from other repositories
CPA-Manager-Plus AGENTS.md
AGENTS.md instructions for seakee/CPA-Manager-Plus, covering repository rules, operation, layout, commands and architecture.
openllmetry CLAUDE.md
Instructions for traceloop/openllmetry, covering openllmetry repository guide, repository structure, nx workspace commands, run tests across all packages and run linting across all packages.
openlit AGENTS.md
AGENTS.md instructions for openlit/openlit, covering openlit agent rules, ce/oss boundary, extension points, project hierarchy and ui and security.
openlit CLAUDE.md
Claude Code instructions for openlit/openlit, a project described as: Open source platform for AI Engineering: OpenTelemetry-native LLM Observability, GPU Monitoring, Guardrails, Evaluations, Prompt Management, Vault, Playground. 🚀💻 Integrates with 50+ LLM Providers, VectorDBs, Agent Frameworks and GPUs.
ai-gateway AGENTS.md
AGENTS.md instructions for ferro-labs/ai-gateway, covering agents.md, project overview, current development snapshot, public-facing wording and build, test, and run commands.
litellm AGENTS.md
AGENTS.md instructions for BerriAI/litellm, a project described as: The fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format with cost tracking, guardrails, load balancing, and logging [Bedrock, Azure, OpenAI, Anthropic, OpenAI, VertexAI, vLLM, Nvidia NIM].