orc-adr-sentinel

orc-adr-sentinel is an agent for coding agents from qGolem/orc. It costs 33 tokens per session (680 once invoked), scanned A, original, MIT.

A background checker that compares project plans with ADRs, written records of important architecture decisions, and the expected roadmap.

In plain words
What is it for?
Use it during phased projects to review PLAN.md against ADR files and ROADMAP.md, then report deviations to the lead.
Why use it?
It catches when a plan conflicts with earlier decisions, introduces an undocumented architecture choice, or moves beyond the planned scope.

Agent

Part of the orc plugin — 29 skills, 27 agents shipped together

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/qgolem/orc/orc-adr-sentinel
Clone the repo
git clone --depth 1 https://github.com/qGolem/orc

Or install orc, the plugin that ships this one along with the rest of its 29 skills, 27 agents.

Wrote 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.

agentmods badge for orc-adr-sentinel

README.md
[![agentmods](https://agentmods.dev/badge/agents/qgolem/orc/orc-adr-sentinel.svg)](https://agentmods.dev/agents/qgolem/orc/orc-adr-sentinel)
Your own site
<a href="https://agentmods.dev/agents/qgolem/orc/orc-adr-sentinel"><img src="https://agentmods.dev/badge/agents/qgolem/orc/orc-adr-sentinel.svg" alt="Measured on agentmods" height="20"></a>
Per session 33 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 680 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00033 $0.00680
Opus 5 $0.00016 $0.00340
Sonnet 5 $0.00007 $0.00136
Haiku 4.5 $0.00003 $0.00068

Measured 5d ago against content hash be734bd11035, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

orc-adr-sentinel 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 5d 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.

agents/orc-adr-sentinel.md · 65 lines

How it starts

The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ADR Sentinel

Persistent teammate that monitors phase plans for ADR compliance and flags architectural drift to the lead asynchronously. Runs in parallel with planner↔checker — does NOT block their convergence loop.

Protocol

  1. Wait for trigger — Planner notifies when PLAN.md is ready (same trigger as checker)
  2. Load ADRsGlob docs/adr/*.md, read metadata + filter by relevance to phase topic
  3. Read PLAN.md — Load the phase plan under review
  4. Check 3 deviation dimensions:
    • Contradiction: Plan decision conflicts with an accepted ADR
    • Coverage gap: Plan introduces architectural choice not covered by any ADR (new ADR needed)
    • Scope drift: Plan approach deviates from what ROADMAP.md anticipated
  5. Report deviations — If found, send structured report to "lead":
    ## ADR Deviation Report — Phase {N}
    
    ### Contradiction
    - PLAN.md: [what the plan says]
    - ADR{X}.{NNN}: [what the ADR says]
    - Suggested resolution: [update ADR / change plan / accept with rationale]
    
    ### Coverage Gap
    - Decision: [architectural choice in plan]
    - No existing ADR covers this
    - Suggested: Run `/orc-adr slug N` to establish ADR
    
    ### Scope Drift
    - ROADMAP.md anticipated: [approach]
    - PLAN.md proposes: [different approach]
    - Suggested resolution: [update ROADMAP / change plan / accept deviation]
    
  6. Stay quiet if clean — No deviations = no message (reduce noise)
  7. Re-check on revision — When planner revises PLAN.md, re-check only previously flagged items
  8. Persist findings — Append all findings to phases/0N-name/DEVIATIONS.md
  9. Shutdown — On shutdown_request, approve immediately

No ADRs Found

If docs/adr/ is empty or does not exist:

  • Skip contradiction check (nothing to contradict)
  • Still run coverage gap + scope drift checks
  • Report: "No ADRs found — consider running /orc-adr slug N to establish architectural decisions."

Key Design Choices

  • Messages go to "lead" (orc-generate orchestrator), NOT to planner — keeps the planner↔checker loop clean
  • Lead decides whether to present deviation to user, ask planner to address it, or accept it
  • Sentinel is async and non-blocking — convergence proceeds regardless of sentinel output
  • Claim "Check ADR compliance for phase {N}" from TaskList when starting

Read the full file on GitHub · 65 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.

  1. 5d ago First seen · 65 lines · 33 tokens per session scan A be734bd11035

Subscribe to this mod's changes

orc-adr-sentinel is an agent published in the GitHub repository qGolem/orc (5 stars, last pushed 5mo ago), licensed MIT. It adds 33 tokens to every session and 680 once invoked, about $0.0002 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.

Related

Other agents, from other repositories

quality-fixer

Specialized agent for verifying software projects and fixing quality failures within the current task scope. Use proactively after code changes or for quality, test, build, lint, format, correctness, or fix requests.

shinpr/claude-code-workflows · 44 tokens

verifier

Critically evaluates investigation results, checks path coverage, and validates failure points using Devil's Advocate method. Use when investigation has completed, or when "verify/validate/double-check/confirm findings" is mentioned. Focuses on verification and conclusion derivation.

shinpr/claude-code-workflows · 54 tokens

code-verifier

Verifies repository-backed claims and implementation feasibility in PRDs, Design Docs, or Work Plans. Use before document review, after implementation, or for reverse-engineered artifact verification.

shinpr/claude-code-workflows · 39 tokens

release-manager

Cuts a brooks-lint release: sets the version in package.json, propagates it across the four plugin manifests and every version-bearing text file via npm run bump, writes the CHANGELOG entry, re-validates, then commits, pushes to main, tags, and publishes the GitHub release. Final pipeline stage of the brooks-harness…

hyhmrright/brooks-lint · 85 tokens

prd-creator

Creates PRD and structures business requirements. Use when new feature/project starts, or when "PRD/requirements definition/user story/what to build" is mentioned. Defines user value and success metrics.

shinpr/claude-code-workflows · 44 tokens

technical-designer

Creates a scoped ADR batch or one backend/general Design Doc from confirmed requirements and decision-relevant repository evidence. Use when technical choices or implementation design need an approved artifact.

shinpr/claude-code-workflows · 37 tokens