Skip to content

docs: the fs report prints, and a reader had no way to learn it stops there - #363

Merged
AminChirazi merged 1 commit into
mainfrom
claude/fs-observation-not-in-trace
Aug 1, 2026
Merged

docs: the fs report prints, and a reader had no way to learn it stops there#363
AminChirazi merged 1 commit into
mainfrom
claude/fs-observation-not-in-trace

Conversation

@AminChirazi

@AminChirazi AminChirazi commented Aug 1, 2026

Copy link
Copy Markdown
Contributor

One paragraph, closing the reader question #360 left open.

The Filesystem observation section says what the report contains and never says where it does not go. "Is this in the trace?" is the first thing anyone with a CI artifact asks, and the answer — no, and deliberately — was recorded on #302 and nowhere a user would ever look.

The distinction worth writing down is that "not implemented yet" and "designed and declined" are different facts, and only one of them tells you to stop waiting. The reason travels with it: a trace is a committed artifact, and these paths are absolute, so /home/alice/exports/acme-corp-2025.csv would be baked into a file that is reviewed and diffed forever — the same argument that keeps execve out of the trap set for its argv.

Closes out #302 entirely. Docs only; ratchets diff size 9 <= 400.

🤖 Generated with Claude Code

… there

The Filesystem observation section described what the report contains and
never said where it does not go. "Is this in the trace?" is the first
question a reader with a CI artifact asks, and the honest answer - no, and
deliberately - was recorded on issue #302 and nowhere a user would look.

Says it, with the reason, because "not implemented yet" and "designed and
declined" are different facts and only one of them tells you to stop waiting.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Updates the agent-testing documentation to explicitly state that the filesystem observation report is intentionally printed only and not persisted into the trace artifact, clarifying an important “designed and declined” product decision for auditors reading CI logs.

Changes:

  • Adds a paragraph explaining that filesystem observation is emitted to stderr and not recorded in the trace.
  • Documents the rationale for not adding an fs trace lane (absolute-path permanence and long-lived reviewability concerns).

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Comment thread docs/agent-testing.md
Comment on lines +941 to +943
survives the run. A lane was designed and declined, because a trace is a
COMMITTED artifact and these paths are absolute - `/home/alice/exports/
acme-corp-2025.csv` would be baked into a file that is reviewed and diffed
@AminChirazi
AminChirazi merged commit effe96a into main Aug 1, 2026
10 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants