Skip to content
Closed
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 3 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -24,14 +24,15 @@ It is intentionally not a hosted scanner. The Action reads the checked-out repos

## Part of an AI-agent governance suite

Four tools mapping orthogonal failure modes of AI-agent deployment:
Five tools mapping orthogonal failure modes of AI-agent deployment:

- **[ScopeTrail](https://github.com/Conalh/ScopeTrail)** — config drift over time (PR-level).
- **[PolicyMesh](https://github.com/Conalh/PolicyMesh)** — policy contradictions across agent surfaces.
- **[CapabilityEcho](https://github.com/Conalh/CapabilityEcho)** — capability drift via code, not config.
- **TaskBound** *(this repo)* — scope creep after the agent runs.
- **[SessionTrail](https://github.com/Conalh/SessionTrail)** — runtime behavior review across agent session transcripts.

The first three are preventive (static analysis of config and code). TaskBound is detective (behavioral, comparing stated intent vs. actual diff).
ScopeTrail, PolicyMesh, and CapabilityEcho are preventive (static analysis of config and code). SessionTrail is runtime (in-session transcript review). TaskBound is detective (stated task vs. actual diff).

## Demo

Expand Down