Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 15 additions & 0 deletions .changeset/00-what-s-new.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
---
"@stainless-code/verdict": minor
---

**What's new in 0.7.0**

Verdict 0.7.0 is a breadth release. It adds a local code index, a read-only `serve` surface with a browser
UI, and a backtest harness that mines past bug fixes from your own history, replays them against your
agent, and scores the result. Review output gains an auditable provenance receipt, diff-coverage
attestation, advisory risk scoring, and an optional validation pass that floors each finding on the
evidence behind it, while the CI gate now checks coverage by default. Posting grows a secret guard, honest
per-host capability reporting, and a summary-comment option, GitLab and Forgejo join GitHub as full publish
targets, and lens packs, declared recipes, and reported verification make the review pipeline itself data.
Thirteen pre-1.0 changes alter behavior — each names the migration step below — and the known limits close
the notes.
25 changes: 25 additions & 0 deletions .changeset/01-new-commands-and-flags.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,25 @@
---
"@stainless-code/verdict": minor
---

**New commands and flags**

- **`pr-wait`** — wait for a PR to merge, close, change checks, or change review state, with a bounded timeout and typed timeout data. [Bounded foreground waits](docs/manual.md#bounded-foreground-waits) <!-- changesets: bounded-foreground-waits.md -->
- **`ingest`** — import SARIF 2.1.0 results into a bundle's findings store, sharing one dedupe and suppression path with model reviews. [Static-analysis ingestion](docs/manual.md#local-review-no-pr) <!-- changesets: ingest-sarif.md -->
- **`threads resolve`** — close the review threads whose findings your store marks addressed, one receipt per run, with `--dry-run` and typed refusals for threads verdict did not author. [Guarded submission](docs/manual.md#guarded-submission-and-operator-surfaces) <!-- changesets: thread-resolve.md -->
- **`risk`** — rank every local bundle into an advisory triage queue by band and 0–100 score, without a host call or an index build. [Triage risk queue](docs/manual.md#triage-risk-queue) <!-- changesets: risk-queue.md -->
- **`report --html`** — export the six dashboard views as a static site, published through staging and backup siblings with rollback and symlink refusals. [Static dashboard export](docs/manual.md#static-dashboard-export) <!-- changesets: report-html.md -->
- **`run --agent`** — run your own agent CLI against an existing bundle, assert exactly one fresh `review.json`, then hand off to the unchanged record/post validation. [Your agent, launched](docs/manual.md#your-agent-launched-run---agent) <!-- changesets: run-agent.md -->
- **`panel`** — plan independent verifier dispatches locally; keep-biased results enrich findings and summaries without voting or dropping anything. [Panel verification](docs/manual.md#panel-verification) <!-- changesets: panel-verification.md -->
- **Recipes** — `.verdict/recipes.json` declares a lens set, synthesizer hint, and budget as strict versioned data, read at the PR base and hashed into the bundle. [Recipes](docs/manual.md#recipes-declared-review-pipelines) <!-- changesets: recipes-format.md -->
- **`skill show`, `skill status`, `skill uninstall`** — inspect one payload file byte-identically, gate on drift with an exit code, and remove only the files a receipt owns. [Install routes](docs/manual.md#install-routes-and-where-commands-run) <!-- changesets: skill-management.md -->
- **`post --summary comment`** — maintain one separate PR-level summary comment, updated in place across all three publish hosts; the default `off` keeps output byte-identical. [Guarded submission](docs/manual.md#guarded-submission-and-operator-surfaces) <!-- changesets: post-summary-comment.md -->
- **`fix --all`** — the batch form of the bounded fix loop: every open inline finding at or above `--min-severity`, one pass per sig under the same ceilings, per-sig outcomes in `data.batch`, and a `fix.patch` written beside the ledger. [Fix loops](docs/manual.md#fix-loops-bounded-verify-gated) <!-- changesets: fix-all-batch.md -->
- **`fix --selected`** — address exactly the sigs a human 👍 selected on the root verdict comment, read through one shared reaction reader; a selection file is written only when something was reacted to. [Fix loops](docs/manual.md#fix-loops-bounded-verify-gated) <!-- changesets: reaction-scoped-fix.md -->
- **`handoff`** — one markdown projection of the open inline findings `fix` would take, selecting through the same selector, with `--json`, `--format md`, and repeatable `--paths`; read-only, with no ledger and no host call. [Handoff](docs/manual.md#handoff-the-fix-brief-for-someone-else) <!-- changesets: handoff-verb.md -->
- **`run --print --stage <n>`** — render one recipe stage at a time, so CI can dispatch the four stages under different models. [Recipes](docs/manual.md#recipes-declared-review-pipelines) <!-- changesets: run-print-stage.md -->
- **`backtest reverts`** — record the reverts git already knows about into a sorted, byte-stable `.verdict/reverts.jsonl`, with `.verdict/reverts.md` as the fenced reading surface. [Reverts](docs/manual.md#reverts) <!-- changesets: backtest-reverts.md -->
- **`backtest mine`** — mine SZZ candidates from a local clone with zero model calls and zero network, writing a timestamp-free `mine.json`. [Backtest](docs/manual.md#backtest-szz-candidates) <!-- changesets: backtest-mine.md -->
- **`backtest replay`** — replay mined candidates against your own agent in a throwaway worktree per candidate, bounded by review count, projected budget, and timeout. [Replay](docs/manual.md#replay) <!-- changesets: backtest-replay.md -->
- **`backtest score`** — grade a replay's findings against the candidate bug lines and report a catch rate, off-target findings per PR, and cost per review as a pure read. [Score](docs/manual.md#score) <!-- changesets: backtest-score.md -->
- **`learn --incident`** — attach one revert or incident to a recorded bundle in a sorted, idempotent `.verdict/incidents.jsonl` ledger. [Knowledge base](docs/manual.md#knowledge-base-scoped-local-rules) <!-- changesets: incident-memory.md -->
15 changes: 15 additions & 0 deletions .changeset/02-local-code-index.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
---
"@stainless-code/verdict": minor
---

**Local code index**

- **`verdict index`** — build a local Go index with the standard library, stored by tree SHA so an empty commit reuses a tree and a rename parses nothing. [Code index](docs/manual.md#code-index) <!-- changesets: index-callers.md -->
- **Graph queries** — `index tests-for`, `index co-changed`, and `index owners` answer candidate tests, co-change partners, and the CODEOWNERS rule at a tree. [Code index](docs/manual.md#code-index) <!-- changesets: index-graph.md -->
- **Three more languages** — the index now reads TypeScript, JavaScript, and Python through two dependency-free lexical scanners, with no tree-sitter and no cgo. [Code index](docs/manual.md#code-index) <!-- changesets: index-scanners.md -->
- **`index blast-radius`** — one change's reach: callers to depth 2 plus candidate tests, co-change partners, and owners for the changed paths. [Code index](docs/manual.md#code-index) <!-- changesets: index-blast-radius.md -->
- **Transitive tests** — `tests-for` walks the reverse import graph up to `--depth` hops and labels tests it reached through imports. [Code index](docs/manual.md#code-index) <!-- changesets: index-transitive-tests.md -->
- **`fetch --index`** — hand an index you already built to `fetch` or `fetch-local` and the bundle gains an `impact` artifact and a fenced `## impact` section. [Impact hand-off](docs/manual.md#impact-hand-off-fetch---index) <!-- changesets: fetch-index-impact.md -->
- **Missing-index hints** — `index blast-radius` and `fetch` now name the revision and tree an absent index is about, with the exact shell-quoted build step that answers for it. [Code index](docs/manual.md#code-index) <!-- changesets: blast-radius-index-hint.md -->
- **Scanner fixes** — four corpus bugs are fixed: a transitive test no longer disappears, relative imports resolve, a Python attribute call is no longer misattributed, and a Go selector outside its package is counted unresolved. [Code index](docs/manual.md#code-index) <!-- changesets: index-corpus-bugs.md -->
- **One impact budget** — the impact section's byte budget is shared across symbols, callers, and tests, each list carrying its own omitted count. [Code index](docs/manual.md#code-index) <!-- changesets: index-impact-budget.md -->
27 changes: 27 additions & 0 deletions .changeset/03-review-quality-and-provenance.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,27 @@
---
"@stainless-code/verdict": minor
---

**Review quality and provenance**

- **Explicit context** — `fetch --context` and `fetch-local --context` attach named context files, and findings cite them. [Explicit context](docs/manual.md#explicit-context) <!-- changesets: context-sources.md -->
- **Coverage basis** — every coverage read now names what its percentage counts: full hunk spans, both sides and context, not just changed lines. [Diff coverage](docs/manual.md#diff-coverage-agent-attestation) <!-- changesets: coverage-scoring-basis.md -->
- **Coverage attestation** — `review.json` gains an optional `coverage[]` checked against the bundle's own hunks, with `--require-coverage` to refuse a partial claim and one body line on a partial attestation. [Diff coverage](docs/manual.md#diff-coverage-agent-attestation) <!-- changesets: diff-coverage-attestation.md -->
- **`ranges: null` is malformed** — a coverage claim whose `ranges` is present but not a string now counts as unattested instead of as the whole file. [Diff coverage](docs/manual.md#diff-coverage-agent-attestation) <!-- changesets: null-ranges-malformed.md -->
- **Ignore overrides** — repeatable `--ignore <glob>` and `--no-default-excludes` on fetch, both recorded in the exclusion receipt. [Exclusions](docs/manual.md#exclusions-and-policy-silence) <!-- changesets: ignore-overrides.md -->
- **Suppression receipts** — excluded diff bytes and withheld findings are receipted, with `post --min-severity` and `report --include-withheld` to read them. [Exclusions](docs/manual.md#exclusions-and-policy-silence) <!-- changesets: ignore-suppression.md -->
- **Proposed learnings** — propose inactive one-rule packs from resolved threads, accept them through the pack importer, or reject them permanently with a reason. [Proposed learnings](docs/manual.md#proposed-learnings) <!-- changesets: learnings-propose.md -->
- **Rule import** — `learn --pack <dir> --id <rule>` imports a structured rule object and keeps its pack provenance and hash. [Import a rule object](docs/manual.md#import-a-rule-object) <!-- changesets: rule-pack-import.md -->
- **Rule exposure** — `learn --status` gains `recalled_in`, `cited_in`, `exposure`, and `acceptance` per rule, reporting unmeasured recall as null rather than zero. [Knowledge base](docs/manual.md#knowledge-base-scoped-local-rules) <!-- changesets: rule-exposure.md -->
- **Risk scoring** — an optional `risk` claim in `review.json` plus `report --risk` and `report --max-risk`, which reports the larger of your claim and a deterministic floor. [Triage risk queue](docs/manual.md#triage-risk-queue) <!-- changesets: risk-score.md -->
- **Reaction proposals** — opt-in reaction proposals from posted comments, plus per-rule acceptance and false-hit rates. [Reaction proposals](docs/manual.md#reaction-proposals-and-rule-rates) <!-- changesets: reactions-trend.md -->
- **Resolution metrics** — live resolution rollups, clean-review metrics, and calibration receipts, with clean reports getting their own headline. [Resolution and clean reviews](docs/manual.md#resolution-and-clean-reviews) <!-- changesets: resolution-rate.md -->
- **Withheld capabilities** — `post.json` records, per capability, what a host could not carry and why, so a degraded publish stays auditable. [Guarded submission](docs/manual.md#guarded-submission-and-operator-surfaces) <!-- changesets: withheld-capabilities.md -->
- **Payload receipts** — a post payload receipt now hashes the rendered review content and the reviewed head, not diff positions or range metadata. [Guarded submission](docs/manual.md#guarded-submission-and-operator-surfaces) <!-- changesets: post-content-digest.md -->
- **Partial publication** — post only the comments and summary parts missing from the fetched remote content, and rearm at a new head with `rearmed` and `previous_head` recorded. [Guarded submission](docs/manual.md#guarded-submission-and-operator-surfaces) <!-- changesets: post-content-recovery.md -->
- **Live authority** — post refreshes live reviews and comments first and refuses without writing when that read fails, so a retry no longer needs a re-fetch. [Guarded submission](docs/manual.md#guarded-submission-and-operator-surfaces) <!-- changesets: post-live-authority.md -->
- **Author fingerprint** — `fetch` writes an advisory `author.json` (`{agent, model, confidence, evidence[]}`) from five harness signals, suggests a reviewer from another model family, and `record.json` carries the author/reviewer pair that `trend` tallies. [Bundle namespace](docs/manual.md#bundle-namespace-core-files-vs-x-) <!-- changesets: author-routing.md, author-routing-hint.md, author-routing-record.md, author-routing-skill.md, author-routing-trend.md -->
- **Revert memory** — a changed path that matches the tracked reverts ledger gives `fetch` a bounded, fenced `reverts.txt`, and the review prompt treats a revert as evidence the change broke something once. [Reverts](docs/manual.md#reverts) <!-- changesets: fetch-reverts.md -->
- **Validation pass** — a trusted `prompts/validate.md` pass writes `validation.json` per sig, `rigor_policy` floors each severity tier, and `post --require-rigor` narrows it per invocation: advisory by default, a receipted withhold in gate mode. [Rigor floors](docs/manual.md#rigor-floors-and-the-validation-pass) <!-- changesets: rigor-validation.md -->
- **Published key names** — `prompts/review.md` tags the evidence rung as `rigor` and the in-bounds flag as `in-bounds`, and an archtest holds every published key to `review.schema.json`. [Rigor floors](docs/manual.md#rigor-floors-and-the-validation-pass) <!-- changesets: prompt-rigor-key.md, prompt-key-shapes.md -->
- **Verdict reconciliation** — binding verdicts reconcile separately from summary and inline content, so a retried approval does not duplicate a GitLab summary note. [Submit per host](docs/manual.md#submit-post-per-host) <!-- changesets: post-verdict-reconciliation.md -->
21 changes: 21 additions & 0 deletions .changeset/04-safety-and-trust.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,21 @@
---
"@stainless-code/verdict": patch
---

**Safety and trust**

- **Tokens off the argv** — Forgejo and Bitbucket reads hand the credential to curl on stdin or a 0600 file, so it is gone from `ps` and `/proc/<pid>/cmdline`. [Providers](docs/manual.md#providers-non-github-hosts) <!-- changesets: curl-token-argv.md -->
- **Secret guard** — `post` refuses a secret-shaped value before the first host call, naming the shape and never the value, with `--allow-secret-values` as the explicit exception. [Guarded submission](docs/manual.md#guarded-submission-and-operator-surfaces) <!-- changesets: post-secret-guard.md -->
- **Mentions neutralized** — `record` escapes `@`-mentions once, where the body is built, so recorded, posted, dry-run, and digested bytes cannot diverge. [Mentions](docs/manual.md#mentions-in-the-published-text) <!-- changesets: mention-neutralization.md -->
- **Punctuated mentions** — `(@alice)`, `**@alice**`, and `_@alice_` are escaped while real addresses stay byte-identical. [Mentions](docs/manual.md#mentions-in-the-published-text) <!-- changesets: punctuated-mentions.md -->
- **Strikethrough mentions** — `~~@alice~~` is escaped too, while a lone tilde in prose stays as written. [Mentions](docs/manual.md#mentions-in-the-published-text) <!-- changesets: strikethrough-mentions.md -->
- **Learning evidence** — cited paths and comment identifiers must stay inside quoted learning evidence, and unsafe ones refuse before a learning is proposed or accepted. [Learnings](docs/manual.md#learnings-durable-repo-memory) <!-- changesets: learnings-evidence.md -->
- **Unicode line breaks** — U+2028 and U+2029 now refuse a serve URL, refuse a learning path, and flatten in a rendered `pack test` line. [Serve](docs/manual.md#serve-local-http-surface) <!-- changesets: line-break-guards.md -->
- **Store locking** — every `.verdict/kb.jsonl` write takes one advisory lock; a second writer refuses `store-busy` instead of dropping a row. [Findings store](docs/manual.md#findings-store--triage-authority) <!-- changesets: store-lock-and-listing.md -->
- **Enforced sandbox** — `verify --sandbox` runs the verify command under bubblewrap on Linux with no network, a private PID namespace, empty `~/.ssh`/`~/.aws`/`~/.docker`, and no silent fallback. [Enforced sandbox](docs/manual.md#enforced-sandbox-verify---sandbox) <!-- changesets: verify-sandbox.md -->
- **macOS sandbox** — `verify --sandbox` now denies every `bind()` and `connect()` on macOS through the built-in `sandbox-exec`, loopback and unix sockets included, so filesystem isolation stays Linux-only. [Enforced sandbox](docs/manual.md#enforced-sandbox-verify---sandbox) <!-- changesets: verify-sandbox-macos.md -->
- **Variable command words refuse** — `E=rm; $E -rf ./target` and the braced, quoted, `export`ed, loop, concatenated, and own-line spellings now refuse `verify-indirection`, including after `env`/`timeout`/`xargs` wrappers and inside `sh -c`. [Known residuals](docs/manual.md#known-residuals-string-scan-ceiling) <!-- changesets: verify-indirection.md -->
- **`.verify.out` is owner-only** — the verify stream is written `0600` on Unix, like `.run.out`, because a command that prints a credential leaves the value in it. [Enforced sandbox](docs/manual.md#enforced-sandbox-verify---sandbox) <!-- changesets: verify-out-owner-only.md -->
- **Launch-record redaction** — a secret-shaped `run --agent` operand no longer reaches the recorded `output_tail` or the 0644 child log. [What `run` asserts](docs/manual.md#what-run-asserts) <!-- changesets: run-argument-credential.md -->
- **setup-verdict hardening** — the action passes its inputs through environment variables and verifies a cached archive before reuse, failing closed on a tampered one. [The runner is your CI](docs/manual.md#the-runner-is-your-ci) <!-- changesets: setup-verdict-hardening.md -->
- **Install preflight** — `skill install` probes exactly the directories it will write before the first byte, so a read-only target leaves nothing half-installed. [Install routes](docs/manual.md#install-routes-and-where-commands-run) <!-- changesets: skill-install-preflight.md -->
11 changes: 11 additions & 0 deletions .changeset/05-agents-and-mcp.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
---
"@stainless-code/verdict": minor
---

**Agents and MCP**

- **Bounded repair loop** — `run --rounds N` re-prompts an agent with the shape checker's fix list, stops on the first valid round, and refuses `repair-exhausted`. [Your agent, launched](docs/manual.md#your-agent-launched-run---agent) <!-- changesets: run-repair-loop.md -->
- **Self-review loop** — a bounded local self-review recipe for Claude, Codex, and OpenCode with round receipts and coverage bound to the verified diff identity. [Agent self-review loop](docs/manual.md#agent-self-review-loop) <!-- changesets: self-review-loop.md -->
- **MCP config for six more hosts** — `mcp config` writes snippets for six more hosts and carries `--read-only` or `--operator` into the argv it prints. [Host config](docs/manual.md#host-config) <!-- changesets: mcp-config-hosts.md -->
- **Declared mutation classes** — every MCP tool states its local mutation class, and a call hidden by `--read-only` returns the typed `read-only-tool` refusal. [MCP server](docs/manual.md#mcp-server-stdio-experimental-until-10) <!-- changesets: mcp-mutation-class.md -->
- **Agent preflight** — a missing agent binary is caught before prompt assembly, with install guidance, and printed or recorded pass-through arguments are limited to safe model and cost options. [Agents, arguments, and `--print`](docs/manual.md#agents-arguments-and---print) <!-- changesets: agent-definition-display.md -->
Loading
Loading