Fix CLI domain failures exiting with code 0

[Nih1tGupta]

Closes #992

Summary

• Propagate non-zero exit codes from Typer/Click when run_cli() uses standalone_mode=False, so contract() / fail() outcomes (e.g. CapabilityNotImplemented, ambiguous_pot, validation errors) reach the OS process instead of always exiting 0.
• Exit with code 1 when graph workbench results emit "ok": false via _emit_graph_result(), closing the gap for in-band failures that never raised an exception (e.g. invalid graph nudge events).

Problem

The CLI printed correct structured errors (JSON or human) but often returned exit code 0, so scripts, CI, and cmd && next_step treated failed commands as successful. Parser/usage errors already exited non-zero; many runtime/domain failures did not.

Root causes:

1. host_cli.run_cli() discarded the return value from app(..., standalone_mode=False). Click returns command exit codes instead of propagating typer.Exit to the process in that mode.
2. _emit_graph_result() printed graph envelopes with "ok": false without raising, so some graph commands exited 0 despite failure payloads.

Solution

File	Change
adapters/inbound/cli/host_cli.py	Capture exit_code = app(...) and raise typer.Exit(code) when non-zero
adapters/inbound/cli/commands/graph.py	After emitting a failed graph envelope, raise typer.Exit(EXIT_VALIDATION) when ok is false

Documented exit contract (unchanged): 0 ok / 1 validation / 2 unavailable / 3 degraded / 4 auth.

Test plan

• potpie --json cloud status → exit 2, code: not_implemented
• potpie --json graph status (ambiguous pot repo) → exit 1, ok: false
• potpie --json graph nudge --event bogus --session s1 --pot <id> → exit 1, ok: false
• potpie pot create (missing name) → exit 2 (parser error, unchanged)
• potpie --json graph catalog --pot <id> → exit 0, ok: true
• potpie --json cloud status && echo fail → chain stops (does not print)
• Unit: test_graph_cli_contract, test_cli_usage_errors, test_sentry_daemon, test_cli_ergonomics, test_cli_bootstrap_status (145 tests)

Manual verification tip: check Potpie’s exit code without piping (cmd; echo $?), since $? after | jq reflects jq, not potpie.

[coderabbitai]

📝 Walkthrough

Summary by CodeRabbit

• Bug Fixes
  • Improved CLI error handling for graph validation failures.
  • Ensured non-zero command results consistently produce the appropriate CLI exit status.
  • Prevented successful-looking command completion when validation errors occur.

Walkthrough

The CLI now propagates graph validation failures and non-zero Typer application return values as process exit codes after emitting available error output.

Changes

CLI Exit Code Propagation

Layer / File(s)	Summary
Graph command failure exit handling potpie/cli/commands/graph.py	_emit_graph_result raises EXIT_VALIDATION when the normalized graph payload has ok: false, after emitting the error output.
CLI exit code capture and propagation potpie/cli/main.py	run_cli captures the Typer app return value and raises typer.Exit with its integer value when it is non-zero.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title matches the main change: fixing CLI failures that were incorrectly exiting with code 0.
Description check	✅ Passed	The description directly describes the same exit-code fix and the affected CLI paths.
Linked Issues check	✅ Passed	The changes implement the linked issue by propagating non-zero CLI exits and raising validation exits for ok:false graph results.
Out of Scope Changes check	✅ Passed	The modified files and behavior are directly tied to the issue and don't introduce unrelated scope.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

📋 Issue Planner

Built with CodeRabbit's Coding Plans for faster development and fewer bugs.

View plan used: #992

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@potpie/context-engine/adapters/inbound/cli/commands/graph.py`:
- Around line 396-398: The downstream ok re-checks are dead code because
_emit_graph_result already raises typer.Exit when payload.get("ok", True) is
False. Remove the redundant post-call checks in _emit_inbox_result,
_emit_quality_result, and _emit_graph_read so those callers just rely on
_emit_graph_result’s exit behavior and no longer repeat unreachable validation
logic.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Repository UI (base), Organization UI (inherited)

Review profile: ASSERTIVE

Plan: Pro

Run ID: 2aa8b21a-dd58-423b-8d78-ab3a1b930c76

📥 Commits

Reviewing files that changed from the base of the PR and between a191a5e and 95e4b30.

📒 Files selected for processing (2)

• potpie/context-engine/adapters/inbound/cli/commands/graph.py
• potpie/context-engine/adapters/inbound/cli/host_cli.py

[coderabbitai]

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

potpie/cli/main.py (1)

94-105: 🎯 Functional Correctness | 🟠 Major | ⚡ Quick win

Route timeline recent through graph failure handling.

timeline_recent enters contract() at Line 694 but ultimately calls _emit_read, which emits an ok: false payload without raising. This bypasses _emit_graph_result and can still return exit code 0 for failed timeline graph reads. Use _graph_command plus _emit_graph_read, or explicitly raise typer.Exit(EXIT_VALIDATION) after emitting a failed result.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@potpie/cli/main.py` around lines 94 - 105, Update timeline_recent so failed
graph reads cannot return exit code 0: route its contract/read flow through
_graph_command and _emit_graph_read, or after emitting a failed result
explicitly raise typer.Exit(EXIT_VALIDATION). Preserve successful timeline
output and ensure the failure path uses the existing graph failure handling.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Outside diff comments:
In `@potpie/cli/main.py`:
- Around line 94-105: Update timeline_recent so failed graph reads cannot return
exit code 0: route its contract/read flow through _graph_command and
_emit_graph_read, or after emitting a failed result explicitly raise
typer.Exit(EXIT_VALIDATION). Preserve successful timeline output and ensure the
failure path uses the existing graph failure handling.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Repository UI (base), Organization UI (inherited)

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 1ec10da4-d30a-4567-8ff0-e3670ea2da0c

📥 Commits

Reviewing files that changed from the base of the PR and between 95e4b30 and 4a54292.

📒 Files selected for processing (2)

• potpie/cli/commands/graph.py
• potpie/cli/main.py