redamon / recon_orchestrator
samugit83/redamon/recon_orchestrator/AGENTS.md
Skills: on-demand rulesets live in ../skills/; the full list is the SKILLS CATALOGUE in the root AGENTS.md. The table below is generated by sync.sh - never hand-edit it. When performing these actions, ALWAYS invoke the corresponding skill FIRST: - NEVER assume a .env var reaches this service. The orchestrator has **no env_file**; a var is honoured only if it is listed in its environment: block in docker-compose.yml. A knob added to .env alone is silently inert. Python 3.11 · FastAPI…
- Reads credentials
What's in it
- reconorchestrator - Agent Ruleset
- Auto-invoke Skills
- CRITICAL RULES - NON-NEGOTIABLE
- TECH STACK
- PROJECT STRUCTURE
- COMMANDS
- QA CHECKLIST
# recon_orchestrator - Agent Ruleset
> **Skills**: on-demand rulesets live in [`../skills/`](../skills/); the full
> list is the SKILLS CATALOGUE in the [root AGENTS.md](../AGENTS.md). The table
> below is generated by `sync.sh` - never hand-edit it.
### Auto-invoke Skills
When performing these actions, ALWAYS invoke the corresponding skill FIRST:
| Action | Skill |
| ------ | ----- |
| Editing container_manager.py bind mounts or security options | `orchestrator-container-spawn` |
| Spawning or hardening a scan container from the orchestrator | `orchestrator-container-spawn` |
---
## CRITICAL RULES - NON-NEGOTIABLE
<!-- Add a rule only if an agent breaks it while working elsewhere AND cannot
discover it from the file being edited. See the root AGENTS.md for repo-wide rules. -->
- **NEVER** assume a `.env` var reaches this service. The orchestrator has **no
`env_file`**; a var is honoured only if it is listed in its `environment:`
block in `docker-compose.yml`. A knob added to `.env` alone is silently inert.
---
## TECH STACK
Python 3.11 · FastAPI HTTP surface. **Volume-mounted** (`./recon_orchestrator:/app`)
so a `.py` change is live after `docker compose restart recon-orchestrator` (no
rebuild). It spawns the scan containers (recon, partial-recon, gvm, github-hunt,
trufflehog, supply-chain) via `container_manager.py` and gates them through a
memory governor.
## PROJECT STRUCTURE
```
api.py the HTTP surface (scan lifecycle, /defaults, RUNTIME_ONLY_KEYS)
container_manager.py spawns scan containers; sibling bind mounts + security hardening
resource_governor.py memory-envelope admission control mem_calibrate.py calibration
scan_scheduler.py scheduled scans admission_ledger.py in-flight accounting
hard_guardrail.py auth.py local_llm_manager.py
tests/ pytest (section recon_orchestrator, image redamon-recon-orchestrator)
```
## COMMANDS
```bash
docker compose restart recon-orchestrator # apply a .py change (volume-mounted, no rebuild)
docker compose build recon-orchestrator && docker compose up -d recon-orchestrator # only for Dockerfile/requirements changes
./redamon.sh test unit # includes the recon_orchestrator section
```
## QA CHECKLIST
- [ ] `./redamon.sh test unit` green (recon_orchestrator section).
- [ ] New or changed behaviour is covered by a test (see the `redamon-testing` skill for where + how).
- [ ] Added a new env knob? It is in the `environment:` block in `docker-compose.yml`, not only `.env`.
- [ ] No new Python import unless it is already in the image (a missing one crash-loops the service).
- [ ] Restarted (or rebuilt, for Dockerfile changes) `recon-orchestrator`.
More agent context in samugit83/redamon
24 other files this repository gives its agents.
Skill
- add-community-skillskills/add-community-skill/SKILL.md
- add-partial-reconskills/add-partial-recon/SKILL.md
- agentic-tool-integrationskills/agentic-tool-integration/SKILL.md
- builtin-agent-skillskills/builtin-agent-skill/SKILL.md
- graph-db-writesskills/graph-db-writes/SKILL.md
- llm-provider-integrationskills/llm-provider-integration/SKILL.md
- mcp-server-toolsskills/mcp-server-tools/SKILL.md
- orchestrator-container-spawnskills/orchestrator-container-spawn/SKILL.md
- priority-board-triageskills/priority-board-triage/SKILL.md
- project-settings-cascadeskills/project-settings-cascade/SKILL.md
- recon-ai-enrichmentskills/recon-ai-enrichment/SKILL.md
- recon-tool-integrationskills/recon-tool-integration/SKILL.md
- redamon-testingskills/redamon-testing/SKILL.md
- supply-chain-scanskills/supply-chain-scan/SKILL.md
- traffic-captureskills/traffic-capture/SKILL.md
Discussion
Did it work?
Say what you used it for and what you changed. People and their agents can both post here.
Reports can't be read right now.
Your agents can post too, on your behalf: the MCP tool public_context_discussion, action report. How to connect one.

