Add Trigger agent cost observability

[ross0x01]

Summary

• tag every initial agent-long Trigger run with permission_full_access or permission_ask_approval
• add Trigger run ID, billed usage duration, and total Trigger cost to hackerai-agent_run
• aggregate approval wait count/wall time plus active model-stream, terminal-wait, and sandbox-recovery durations
• preserve zero-valued telemetry and cover the wiring with regression tests

Validation

• jest --runInBand --no-cache lib/api/__tests__/chat-logger.test.ts lib/chat/__tests__/agent-run-timing.test.ts lib/api/__tests__/agent-long-contracts.test.ts lib/ai/tools/__tests__/run-terminal-cmd.test.ts lib/ai/tools/__tests__/interact-terminal-session.test.ts (144 tests passed)
• tsc --noEmit --incremental false
• changed-file ESLint
• full ESLint (0 errors; 6 existing warnings)
• Prettier check and git diff --check

Summary by CodeRabbit

• New Features

  • Added detailed agent-run timing and cost telemetry, including model streaming, terminal waits, sandbox recovery, and approval waits.
  • Agent runs now include permission-mode metadata for improved tracking.
  • Analytics capture expanded trigger usage, cost, timing, and recovery metrics.

• Bug Fixes

  • Ensured timing metrics are recorded even when terminal operations or other measured tasks fail.
  • Improved sandbox recovery tracking and readiness verification.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
hackerai	Ready	Preview, Comment	Jul 15, 2026 9:49pm

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: b6d87d66-176a-4239-812b-4558163c765b

📥 Commits

Reviewing files that changed from the base of the PR and between 5d2a9a4 and f0079ec.

📒 Files selected for processing (12)

• lib/ai/tools/index.ts
• lib/ai/tools/interact-terminal-session.ts
• lib/ai/tools/run-terminal-cmd.ts
• lib/api/__tests__/agent-long-contracts.test.ts
• lib/api/__tests__/chat-logger.test.ts
• lib/api/agent-stream-runner.ts
• lib/api/agent-trigger-route.ts
• lib/api/chat-logger.ts
• lib/chat/__tests__/agent-run-timing.test.ts
• lib/chat/agent-run-timing.ts
• trigger/agent-long.ts
• types/agent.ts

---

📝 Walkthrough

Walkthrough

Agent runs now track approval waits, model streaming, terminal waits, sandbox recovery, Trigger usage, and cost. The telemetry flows through tool and stream contexts into completion analytics, while Trigger tags include the requested permission mode.

Changes

Agent Runtime Telemetry

Layer / File(s)	Summary
Timing contracts and tracker types/agent.ts, lib/chat/agent-run-timing.ts, lib/chat/__tests__/agent-run-timing.test.ts	Adds active-time categories, timing snapshots, and AgentRunTimingTracker coverage for approval waits, model streams, measured operations, and error handling.
Execution timing instrumentation lib/ai/tools/index.ts, lib/ai/tools/interact-terminal-session.ts, lib/ai/tools/run-terminal-cmd.ts, lib/api/agent-stream-runner.ts, trigger/agent-long.ts, lib/api/__tests__/agent-long-contracts.test.ts	Wraps terminal waits and sandbox recovery, connects approval-wait tracking, and adds model stream lifecycle callbacks to the agent execution flow.
Completion telemetry and analytics trigger/agent-long.ts, lib/api/chat-logger.ts, lib/api/__tests__/chat-logger.test.ts, lib/api/__tests__/agent-long-contracts.test.ts	Combines Trigger usage with timing snapshots and forwards trigger duration, cost, approval, streaming, terminal, and recovery metrics to hackerai-agent_run.
Trigger permission metadata lib/api/agent-trigger-route.ts, lib/api/__tests__/agent-long-contracts.test.ts	Adds a permission-mode tag to triggered agent runs and verifies queued trigger metadata and status fields.

Estimated code review effort: 4 (Complex) | ~45 minutes

Sequence Diagram(s)

sequenceDiagram
  participant AgentRun
  participant AgentStreamRunner
  participant TerminalTools
  participant TimingTracker
  participant Analytics
  AgentRun->>AgentStreamRunner: start and finish model stream timing
  AgentRun->>TerminalTools: provide active-time measurer
  TerminalTools->>TimingTracker: record terminal wait or sandbox recovery
  AgentRun->>TimingTracker: record approval waits
  AgentRun->>Analytics: submit Trigger usage and timing snapshot
  Analytics->>Analytics: emit hackerai-agent_run properties

Loading

Possibly related PRs

• hackerai-tech/hackerai#882: Extends the same agent completion analytics plumbing with additional hackerai-agent_run telemetry fields.
• hackerai-tech/hackerai#895: Modifies the same chat-logger.ts agent analytics event emission paths.
• hackerai-tech/hackerai#826: Touches the shared agent tool wiring and terminal execution paths used by the new timing wrappers.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly matches the PR’s main focus on Trigger agent cost and telemetry observability.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/trigger-agent-cost-observability

---

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