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 agents/mohammad19974/bounded-relay/handoffgit clone --depth 1 https://github.com/mohammad19974/bounded-relayWhat 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 | $0.00969 |
| Opus 5 | $0.00000 | $0.00485 |
| Sonnet 5 | $0.00000 | $0.00194 |
| Haiku 4.5 | $0.00000 | $0.00097 |
Grade A, and why
HANDOFF 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.
How it starts
The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BoundedRelay Handoff
State: verified locally — publication and external CI remain user-owned
Objective
Publish-ready implementation of Portable Policy Profiles at
specs/002-portable-policy-profiles without changing the established one-way
Claude Code host -> BoundedRelay MCP -> local Codex CLI topology or the legacy
v0.1 routing fingerprint.
Active artifacts
specs/002-portable-policy-profiles/spec.mdspecs/002-portable-policy-profiles/plan.mdspecs/002-portable-policy-profiles/tasks.mddocs/project-profiles.mddocs/adr/0010-portable-intersection-only-project-profiles.md.specify/memory/constitution.md.specify/agents/context.json
Completed state
- T001-T024 are complete. The feature adds strict non-executable project profiles, canonical fingerprints, deterministic capability routing, narrower write policy, required check bindings, explicit Codex-only model policy, and a plan-level cross-review policy.
- The legacy no-profile path remains
sdd-routing-v2/sdd-task-fit-v1; the opt-in profile path issdd-routing-v3/sdd-capability-fit-v1. - Profiles never choose or launch a Claude model, execute declared argv, expand server authority, apply patches, commit, merge, push, publish, or deploy.
- Spec Kit replays the authoritative route, validates tree-bound receipts, rejects more than 256 required writer receipts before execution, and rejects any active wave that would raise all recorded writer receipts above 256 after optional receipts are included.
- Doctor capability detection uses bounded raw probe text internally and exposes only separately redacted diagnostics, including when a forwarded environment value equals a required Codex flag.
- CI defines Node 22.13 and 24 checks across Linux, macOS, and Windows, plus an installed-package contract and platform-specific npm shim verification.
- Independent final review found no remaining Critical, High, or Medium release blocker after all accepted findings were repaired.
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 · 89 lines · 0 tokens per session scan A 6cd7af0f06d6
HANDOFF is an agent published in the GitHub repository mohammad19974/bounded-relay (1 stars, last pushed 3d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 969 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 2026-08-31.
Other agents, from other repositories
issue-tracker
Issues and PRDs for this repository live in GitHub Issues. Use the gh CLI from this clone so the repository is inferred from origin.
triage-labels
All canonical role labels are configured in GitHub.
cv-problem-solver
Computer-vision problem solver. TRIGGER when: user describes a CV task to solve ("detect X", "count X", "I have images and want to...", "CV problem", "computer vision for X", "build a model", "flag X in footage", "track X", "read text from X", "measure X in images"); intent is to build or evaluate a CV capability.…
economics-consultant
Computer-vision economics consultant. TRIGGER when: user invokes $estimate-economics in Codex or /sentinel:estimate-economics in Claude Code, asks for annotation or labeling cost, training cost, managed vs self-hosted cost, build-vs-buy, scale economics, deployment crossover, or selected the "managed at scale" /…
explore
Semantic codebase exploration using Context King protocol.
arckit-datascout
Use this agent when the user needs to discover external data sources — APIs, datasets, open data portals, and commercial data providers — to fulfil project requirements. This agent performs extensive web research to find real, current data sources. Examples: Context: User has a project with requirements and wants to…