chore(deps): update dependency @earendil-works/pi-coding-agent to v0.84.3

[renovate-fullsend]

This PR contains the following updates:

Package	Change	Age	Confidence
@earendil-works/pi-coding-agent (source)	0.84.2 → 0.84.3

---

Release Notes

earendil-works/pi (@​earendil-works/pi-coding-agent)

v0.84.3

Compare Source

New Features

• PowerShell tool — Use optional native PowerShell command execution on Windows. See PowerShell Tool.
• Safer managed updates — Stage, verify, and atomically activate updates for installer-managed installations. See Install and Manage.
• Model and thinking controls — Select thinking levels with /thinking, search defaults, keep selections session-scoped, and persist them explicitly with Ctrl+S. See Models and Thinking.

Breaking Changes

• Renamed the inherited GoogleThinkingLevel type to GoogleApiThinkingLevel and added ResolvedGoogleThinkingLevel for normalized adapter levels.

Added

• Added an optional powershell tool for Windows, configurable through defaultTools and the SDK. See PowerShell Tool.
• Added a /thinking selector and searchable default choices to the model and thinking selectors; Ctrl+S saves the selected model as the global default. See Models and Thinking.
• Added optional routing session IDs to exported compaction summary helpers so callers can preserve provider routing without enabling prompt cache writes.
• Added transcript usage notices for compaction and branch summaries when cache miss notices are enabled.
• Added session_compact_failed extension events so compaction failures and aborts expose their reason, retry state, source, and error message to handlers (#​8175).
• Added inherited provider-neutral toolChoice support to simple stream requests.
• Added inherited automatic Anthropic server-side refusal fallback for supported first-party models, including returned-model usage pricing (#​8017).
• Added inherited configurable OpenAI-compatible thinking-token budget fields for vLLM, Qwen/SGLang, and llama.cpp servers. See OpenAI Compatibility (#​8275 by @​bnsd55).
• Added inherited China-specific ZAI Coding Plan models, including GLM-4.6V vision support and API-equivalent usage cost estimates (#​8220).
• Added inherited deepseek-v4-pro-0813 support to the Qwen Token Plan Individual catalog (#​8194).

Changed

• Changed experimental installer-managed installations so pi update stages, verifies, and atomically activates the selected release in place. See Install and Manage.
• Changed inherited built-in xAI models to use the Responses API with encrypted reasoning replay and made Grok 4.6 the default xAI model (#​8124 by @​Jaaneek).
• Changed inherited Anthropic, Azure OpenAI, Google, Mistral, and OpenAI adapters to send Pi's default User-Agent unless overridden (#​8305).
• Changed Windows and WSL keybinding defaults to avoid terminal-reserved shortcuts for image paste, model cycling, editor undo, fullscreen transcript navigation and search, and message queueing (#​8372).
• Changed Bun release archives to ship the native clipboard binary only inside the wrapper package, removing a duplicate platform package from each archive.
• Changed package resource glob expansion to use Node.js's built-in implementation with deterministic visible-path matching, reducing the installed runtime dependency tree.
• Changed the bundled Node.js runtime to load jiti only when importing an extension and Babel only when uncached source needs transformation, reducing CLI startup time and bundle size.
• Changed syntax highlighting to initialize only twenty common languages eagerly and defer the remaining grammars until after the initial TUI render, reducing CLI startup time.
• Changed the Node.js CLI and RPC entrypoints to load a bundled runtime, reducing startup filesystem reads while keeping the public library and legacy module paths on the modular runtime for normal dependency identity.
• Changed session sharing to render clickable terminal links, display only the canonical Radius artifact URL, and include the current system prompt and active tool definitions in Radius session shares.

Fixed

• Fixed failed extension factories leaving event subscriptions, provider registrations, and default flag state active (#​8424 by @​acmerfight).
• Fixed models.json typings omitting the documented OpenAI-compatible compat.supportsFinishReason provider and model override (#​8487 by @​petrroll).
• Fixed /model and /thinking selections being persisted globally unless explicitly saved with Ctrl+S (#​5263).
• Fixed JSON and RPC toolcall_start events omitting the tool call id and name (#​7953 by @​christianklotz).
• Fixed extensions failing to load when the Node.js CLI runs as a single-executable application (#​8237).
• Fixed nested Markdown skills inside .agents/skills/ grouping directories not being discovered.
• Fixed compaction and branch summarization requests exposing tools to providers.
• Fixed single-object edit tool inputs failing validation by accepting them as one-edit arrays in both coding-agent and harness edit tools (#​7835).
• Fixed root Markdown files such as README.md and AGENTS.md in skill directories being reported as broken skills unless they declare valid skill frontmatter (#​7805).
• Fixed the default Cerebras model referencing an unavailable Z.AI model.
• Fixed inherited OpenAI-compatible Chat Completions reasoning replay to preserve and resend assistant-level reasoning_details verbatim and in order (#​7994).
• Fixed inherited Anthropic server-side fallback responses being priced with the requested model instead of the returned fallback model (#​8285).
• Fixed inherited GitHub Copilot login triggering model-policy rate limits by limiting policy updates, retrying model discovery once, and honoring server retry delays (#​7850).
• Fixed inherited Amazon Bedrock dropping and failing to replay opaque redacted reasoning from non-Anthropic models (#​8314 by @​seiji).
• Fixed inherited Z.AI Coding Plan models deriving incomplete reasoning-effort metadata, including missing GLM-5.3 low, high, and max levels (#​8336).
• Fixed inherited DeepSeek V4 Flash on OpenCode and OpenCode Go omitting its supported low thinking level (#​8181 by @​tianshuang).
• Fixed inherited Azure OpenAI Responses ignoring toolChoice in provider-specific stream requests.
• Fixed inherited Amazon Bedrock response hooks receiving only a synthesized request id instead of the raw response headers (#​8234).
• Fixed inherited Kimi usage reporting so top-level cached_tokens count as cache reads instead of normal input tokens (#​8075).
• Fixed inherited Google custom models ignoring thinkingLevelMap, which dropped extended thinking controls (#​8135).
• Fixed writes to auth.json and models-store.json overriding administrator-managed file permissions and ACLs (#​7779).
• Fixed UTF-8 BOM markers preventing frontmatter and user configuration files from loading (#​8337).
• Fixed invalid settings files being easy to miss during interactive startup by rendering warnings with the file path inside the TUI (#​7829).
• Fixed the subagent example repeatedly prompting before running project-local agents in trusted repositories (#​8261).
• Added session_compact_failed extension events so compaction failures and aborts expose their reason, retry state, source, and error message to handlers (#​8175).
• Fixed truncated compaction and branch summaries being persisted when generation reaches its output token limit (#​7048).
• Fixed npm package update checks treating older registry versions as available updates, preventing pi update from downgrading already-newer installed packages (#​8226).
• Fixed built-in llama.cpp models disappearing from /model when /llama refreshed a configured server under PI_OFFLINE, and included idle-slept sleeping router models plus autoloadable unloaded presets in the selectable catalog (#​8167).
• Fixed pi.registerFlag() accepting default values that do not match the declared flag type (#​8064).
• Fixed Z.AI Coding Plan defaults referencing the removed GLM-5.1 model (#​8096).
• Fixed repeated ambiguous truncated-response recovery being mislabeled as context overflow (#​8130).
• Fixed duplicate fullscreen right-click paste in VS Code-based terminals on Windows (#​8186).
• Fixed inherited padded text exceeding narrow terminal widths (#​8252).
• Fixed inherited wrapped Markdown table links leaking color into borders and neighboring cells, including tables inside blockquotes (#​8335).
• Fixed llama.cpp login guidance to direct users to /llama before /model when no local models are loaded (#​8203).
• Fixed hung pi.dev model catalog requests consuming the entire refresh deadline without retrying (#​8198).
• Fixed inherited Xiaomi model catalogs listing shut-down MiMo V2 models in /model and --list-models (#​8187).
• Fixed branch summary entries recording the navigation destination in fromId instead of the pre-navigation source leaf.
• Fixed threshold auto-compaction being skipped when providers omit streaming usage data (#​8328).
• Fixed dash-prefixed prompts being parsed as options by supporting -- as an end-of-options delimiter (#​7269).

---

Configuration

📅 Schedule: (UTC)

• Branch creation
  • At any time (no schedule defined)
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR is behind base branch, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box

---

This PR has been generated by Mend Renovate.

[fullsend-ai-review]

🤖 Finished Review · ✅ Success · Started 3:27 PM UTC · Completed 3:37 PM UTC

Commit: 91a0e61 · View workflow run →

Runtime: claude · Model: opus → claude-opus-4-6 · Effort: high · Cost: $1.39

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[fullsend-ai-review]

Review

Findings

High

• [protected-path] images/sandbox/Containerfile:68 — This PR modifies a file under the protected path images/. No linked issue authorizes the change to this governance/infrastructure file. While this is an automated Renovate dependency update (bumping @earendil-works/pi-coding-agent from 0.84.2 to 0.84.3) and the Containerfile documents Renovate authorization for this ARG, human approval is always required for protected-path changes.
  Remediation: A human reviewer must approve this change.

Medium

• [consumer-completeness] images/sandbox/Containerfile:68 — The Containerfile comment explicitly requires re-verification of the pi-anthropic-vertex extension's compatibility on every PI_VERSION bump (To update (also re-run on every PI_VERSION bump)). The extension at v0.1.13 was synced against pi 0.81.1, and PI_VERSION is now 0.84.3. The v0.84.3 release notes include a breaking change in the thinking/model surface area (GoogleThinkingLevel renamed to GoogleApiThinkingLevel) — the area the extension mirrors. There is no evidence in this PR that the extension's compatibility was re-checked.
  Remediation: Verify the pi-anthropic-vertex extension (v0.1.13) works correctly with pi 0.84.3 by checking the extension's sync/compat.json upstream and confirming the @anthropic-ai/sdk version override in its package.json still matches pi 0.84.3's packages/ai/package.json. If the extension needs updating, bump PI_ANTHROPIC_VERTEX_VERSION and refresh the SHA256.

Low

• [stale-reference] internal/runtime/pi_progress.go:12 — Multiple code comments and test fixtures reference v0.84.2 as the wire-format version the parser was built against (pi_progress.go, pi_run.go, regen.sh). With PI_VERSION now at 0.84.3, these references are stale. The Containerfile comment warns that the pi stream parser fixtures must be re-verified when this pin moves.
  Remediation: Re-generate the pi stream parser fixtures using internal/runtime/testdata/pi/regen.sh and update the version references in code comments.

---

Next steps:

• /fs-fix — agent addresses review findings automatically
• /fs-fix <your instruction> — agent fixes with your specific guidance
• Push commits directly — review re-runs automatically on push
• /fs-fix-stop — disable automatic fix runs for this PR

Previous run

Review

Findings

High

• [protected-path] images/sandbox/Containerfile:68 — This PR modifies a file under the protected path images/. No linked issue authorizes the change to this governance/infrastructure file. While this is an automated Renovate dependency update (bumping @earendil-works/pi-coding-agent from 0.84.2 to 0.84.3) and the Containerfile documents Renovate authorization for this ARG, human approval is always required for protected-path changes.
  Remediation: A human reviewer must approve this change. Additionally, verify that the pi stream parser fixtures have been re-verified per the Containerfile comment ("the pi stream parser fixtures must be re-verified when this pin moves") and confirm pi-anthropic-vertex extension (v0.1.13, synced against pi 0.81.1) compatibility with pi 0.84.3.

---

Next steps:

• /fs-fix — agent addresses review findings automatically
• /fs-fix <your instruction> — agent fixes with your specific guidance
• Push commits directly — review re-runs automatically on push
• /fs-fix-stop — disable automatic fix runs for this PR

[fullsend-ai-review]

🤖 Finished Review · ✅ Success · Started 3:31 AM UTC · Completed 3:46 AM UTC

Commit: 635e9d5 · View workflow run →

Runtime: claude · Model: opus → claude-opus-4-6 · Effort: high · Cost: $2.79

[fullsend-ai-review]

See the review comment for full details.

[fullsend-ai-retro]

🤖 Finished Retro · ✅ Success · Started 9:29 AM UTC · Completed 9:47 AM UTC

Commit: 635e9d5 · View workflow run →

Runtime: claude · Model: opus → claude-opus-4-6 · Effort: high · Cost: $5.07

[fullsend-ai-retro]

Retro: PR #6552 — pi-coding-agent v0.84.2 → v0.84.3

A Renovate bot opened this 1-line patch bump in images/sandbox/Containerfile. The review agent ran twice ($1.39 + $2.79 = $4.18 total), submitted CHANGES_REQUESTED both times, triggered two Fix agent dispatches that both failed at the eligibility gate (Renovate is not the coder bot), and the PR was ultimately approved and merged by a human reviewer without comment.

Timeline

1. Aug 24 15:25 — Renovate opens PR, CI and review agent dispatched.
2. Aug 24 15:27 — Review run 1 completes (24 turns, 22 tool calls, $1.39). Flags [HIGH] protected-path.
3. Aug 24 15:37 — Review bot submits CHANGES_REQUESTED → triggers Fix dispatch.
4. Aug 24 15:38 — Fix dispatch fails: Renovate is not the coder bot and PR lacks fullsend-fix label.
5. Aug 25 03:30 — Renovate rebases (force-push, new commit 635e9d5). Second review dispatched.
6. Aug 25 03:32 — Review run 2 completes (28 turns, 60 tool calls, $2.79). Force-push caused changed_since_prior = all, triggering full-scope re-review with correctness sub-agent (25 tool calls) and challenger (9 tool calls). Finds 3 issues: [HIGH] protected-path, [MEDIUM] consumer-completeness, [LOW] stale-reference.
7. Aug 25 03:47 — Review bot submits CHANGES_REQUESTED again → second Fix dispatch fails same way.
8. Aug 25 09:19 — Human (rh-hemartin) approves without comment.
9. Aug 25 09:28 — PR merged.

Review quality assessment

• [HIGH] protected-path — Accurate and appropriate. images/ is in REVIEW_PROTECTED_PATHS and the renovate.json config explicitly sets automerge: false for this package to require human review of stream parser fixtures. The gate worked correctly.
• [MEDIUM] consumer-completeness — Valid concern (extension is 3 minor versions behind its compat range), but the specific technical claim was wrong: the agent asserted the GoogleThinkingLevel → GoogleApiThinkingLevel rename affects pi-anthropic-vertex, but the extension never imports either type. The challenger sub-agent did not catch this factual error.
• [LOW] stale-reference — Accurate that version references exist; appropriately low severity for a patch bump.

The human approval was reasonable — this is a patch bump with CI passing and the reviewer made a judgment call on risk level.

Waste analysis

Total infrastructure cost for a 1-line version bump: $4.18 in review, 2 wasted Fix dispatches, 6 fullsend.yaml routing runs, duplicate CI/E2E suites (triggered by the Renovate rebase). The style-conventions sub-agent also failed silently in both runs due to model unavailability.

Evidence for existing issues

All findings from this retro map to existing open issues — no new proposals warranted:

• Review using CHANGES_REQUESTED on bot PRs triggers unfixable Fix dispatches: This PR produced 2 wasted Fix dispatches. Supports fullsend#5369 (use COMMENT instead of CHANGES_REQUESTED for governance-only findings on bot PRs) and fullsend#5811 (skip fix dispatch at routing level for ineligible PRs).
• Renovate rebase triggered expensive full-scope re-review: The force-push caused changed_since_prior = all despite identical diff, doubling review cost. Supports fullsend#4401 (detect rebase-only force-pushes) and fullsend#1355 (skip re-review for bot dep-bump rebases with no semantic change).
• Challenger failed to catch hallucinated dependency-impact claim: The correctness sub-agent fabricated a technical connection (GoogleThinkingLevel rename → pi-anthropic-vertex extension) that doesn't exist in the code, and the challenger didn't verify it. Supports agents#832 (unsubstantiated verification claims) and agents#420 (verify technical assertions against source code before posting).
• Style-conventions sub-agent consistently failing: Failed in both runs due to model unavailability. Supports agents#379 (model unavailability fallback) and agents#256 (model alias mismatch).