Skip to content

Agent-estimate

v0.7.1 Bugfix

This release fixes issues for SREs watching stability and regressions.

Published 3d AI Agents & Assistants
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

ai-agents claude-code cli developer-tools effort-estimation estimation
+5 more
metr multi-agent pert python wave-planning

ReleasePort's take

Light signal
editorial:auto 3d

Release v0.7.1 fixes six engine correctness gaps and updates documentation, README examples, plugin skill exposure, and removes unused slash commands.

Why it matters: Addresses six critical correctness bugs (severity 40) across CLI flag processing, METR agent logic, timestamp parsing, calibration, wave planning, and deprecates two undocumented slash‑command APIs; operators should review the changes before upgrading.

Summary

AI summary

Fixed six engine correctness gaps and updated documentation, README examples, plugin skill exposure, and removed unused slash commands.

Changes in this release

Feature Low

Restores `/estimate` skill in Claude Code plugin via updated `plugin.json` path.

Restores `/estimate` skill in Claude Code plugin via updated `plugin.json` path.

Source: llm_adapter@2026-06-11

Confidence: high

 —
Deprecation Low

Drops undocumented `/validate-estimate` and `/calibrate` slash commands; routes validation/calibration through `/estimate validate` and `/estimate calibrate`.

Drops undocumented `/validate-estimate` and `/calibrate` slash commands; routes validation/calibration through `/estimate validate` and `/estimate calibrate`.

Source: llm_adapter@2026-06-11

Confidence: high

 —
Bugfix Medium

Fixes early `--format` validation before side effects.

Fixes early `--format` validation before side effects.

Source: llm_adapter@2026-06-11

Confidence: high

 —
Bugfix Medium

Fixes explicit `--warm-context 1.0` handling.

Fixes explicit `--warm-context 1.0` handling.

Source: llm_adapter@2026-06-11

Confidence: high

 —
Bugfix Medium

Fixes friction‑aware assigned‑agent METR re‑checks.

Fixes friction‑aware assigned‑agent METR re‑checks.

Source: llm_adapter@2026-06-11

Confidence: high

 —
Bugfix Medium

Fixes Python 3.10‑compatible Z timestamp parsing.

Fixes Python 3.10‑compatible Z timestamp parsing.

Source: llm_adapter@2026-06-11

Confidence: high

 —
Bugfix Medium

Fixes UTC‑normalized calibration cohorts/order.

Fixes UTC‑normalized calibration cohorts/order.

Source: llm_adapter@2026-06-11

Confidence: high

 —
Bugfix Medium

Fixes `plan_waves` duplicate/negative input validation.

Fixes `plan_waves` duplicate/negative input validation.

Source: llm_adapter@2026-06-11

Confidence: low

 —
Bugfix Low

Updates README hero example to use real CLI surface with captured output.

Updates README hero example to use real CLI surface with captured output.

Source: llm_adapter@2026-06-11

Confidence: high

 —
Bugfix Low

Grants `issues: read` permission in README Action workflow example for private repos.

Grants `issues: read` permission in README Action workflow example for private repos.

Source: llm_adapter@2026-06-11

Confidence: high

 —
Bugfix Low

Validates `plan_waves` inputs to reject duplicates/negatives.

Validates `plan_waves` inputs to reject duplicates/negatives.

Source: granite4.1:30b@2026-06-11-audit

Confidence: low

 —
Bugfix Low

Updates METR model keys in `examples/multi-agent.md` output to current versions.

Updates METR model keys in `examples/multi-agent.md` output to current versions.

Source: granite4.1:30b@2026-06-11-audit

Confidence: low

 —
Full changelog

Fixed

  • Fixed six engine correctness gaps: early --format validation before side effects, explicit --warm-context 1.0 handling, friction-aware assigned-agent METR re-checks, Python 3.10-compatible Z timestamp parsing, UTC-normalized calibration cohorts/order, and plan_waves duplicate/negative input validation (#31, #33, #36, #37, #38, #39).
  • README hero example now uses the real CLI surface with real captured output — the previous block invoked a nonexistent --model flag, passed an unsupported second positional task, and showed fabricated numbers (#28).
  • README Action workflow example now grants issues: read, required for issue fetching on private repos where an explicit permissions: block zeroes unlisted scopes (#40).
  • Claude Code plugin exposes the /estimate skill again: plugin.json now points skill discovery at skills/estimate/claude/, which the v0.7.0 multi-runtime restructure had moved out of the default skills/<name>/SKILL.md scan path (#29).
  • Dropped the phantom /validate-estimate and /calibrate slash commands (documented but never installable); validation and calibration now route through the installed skill as /estimate validate and /estimate calibrate (#29).
  • Refreshed stale METR model keys in examples/multi-agent.md output (gpt_5_3 → gpt_5_4, gemini_3_pro → gemini_3_1_pro) to match current output (#28).

Full Changelog: https://github.com/kiloloop/agent-estimate/compare/v0.7.0...v0.7.1

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track Agent-estimate

Get notified when new releases ship.

Sign up free

About Agent-estimate

All releases →

Related context

Beta — feedback welcome: [email protected]