zsp/everything-claude-code
1
0
Fork 0
mirror of https://github.com/affaan-m/everything-claude-code.git synced 2026-06-12 19:23:07 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: stability batch — hook stdin truncation, Codex exa TOML, Stop hook JSON, GateGuard repetition (#2227)

Browse Source 
* fix(hooks): fail open on oversized stdin instead of echoing truncated JSON (#2222)

run-with-flags.js capped stdin at 1MB but every fallthrough path still
echoed the truncated string to stdout. The harness parses hook stdout as
JSON, got a document cut mid-stream, and blocked the tool call — so any
Edit/Write with a >1MB hook payload was permanently blocked by every
registered pre-write hook, before ECC_HOOK_PROFILE / ECC_DISABLED_HOOKS
gating could run.

- Exit 0 with empty stdout (no opinion) when the stdin cap trips, before
  any echo or gating logic.
- Flush stdout via write callback before process.exit: exiting right
  after stdout.write() dropped everything past the ~64KB pipe buffer,
  cutting even sub-cap pass-through payloads mid-JSON.

Regression tests cover the enabled, disabled, and missing-arg paths for
oversized payloads plus full echo of sub-cap >64KB payloads.

* fix(codex): stop emitting invalid exa url entry, align merge with connector policy (#2224)

The Codex MCP merge declared exa with a url key, but Codex's
[mcp_servers.*] TOML schema is stdio-only — the url key makes the
entire config.toml fail to load, bricking both the codex CLI and the
desktop app. Every install/update re-injected the line because the
urlEntry branch treated the broken entry as present.

- ECC_SERVERS now emits only the current default set per
  docs/MCP-CONNECTOR-POLICY.md: chrome-devtools (stdio, command/args).
  Retired servers (supabase, playwright, context7, exa, github, memory,
  sequential-thinking) are never re-emitted; existing user-managed
  entries are untouched.
- The merge now repairs the exact ECC-emitted broken form (url-only
  exa entry) on every run so re-running the installer fixes broken
  configs instead of preserving them. User stdio exa entries
  (command + mcp-remote) are left alone.
- check-codex-global-state.sh requires chrome-devtools instead of the
  retired set, and flags url-only exa entries with a repair hint.

Tests cover repair, re-run idempotence, stdio-entry preservation, and
no-retired-server emission in add, update, dry-run, and disabled modes.

* fix(hooks): never echo truncated stdin from Stop hooks (#2090)

Stop hooks follow the ECC pass-through convention (echo stdin on
stdout), but every echoing Stop hook capped stdin and echoed the capped
string. The Stop payload carries last_assistant_message, so a long
final assistant message produced a JSON document cut mid-stream on
stdout, which the harness reports as 'Stop hook error: JSON validation
failed' across the whole Stop chain.

Reproduced: a Stop payload with a >64KB last_assistant_message run
through run-with-flags + cost-tracker emitted exactly 65536 bytes of
invalid JSON (cost-tracker capped stdin at 64KB — far below realistic
Stop payloads).

- cost-tracker: raise the cap to 1MB (matching all other hooks) and
  suppress the pass-through echo when stdin was truncated.
- check-console-log, stop-format-typecheck, desktop-notify: suppress
  the echo when stdin was truncated; flush stdout before process.exit
  so sub-cap payloads are not cut at the ~64KB pipe buffer.
- All hooks keep exiting 0 (fail-open); diagnostics go to stderr.

New stop-hooks-stdout test asserts the contract for every registered
Stop hook: stdout is empty or valid JSON, exit code 0 — for realistic
100KB payloads and oversized >1MB payloads, via the production runner
and via direct invocation. Updated the old hooks.test.js case that
codified the truncated-echo behavior.

* fix(hooks): dampen GateGuard fact-force repetition in long sessions (#2142)

In long autonomous sessions the fact-force gate produced 10+
near-identical 'state facts -> blocked -> restate -> retry' blocks in
one context window, which measurably raises the odds of the model
collapsing into a degenerate single-token repetition loop.

- Track a per-session fact_force_denials counter in GateGuard state
  (merged max across concurrent writers, reset with the session, robust
  to malformed on-disk values).
- The first GATEGUARD_FACT_FORCE_FULL_DENIALS denials (default 3) keep
  the full four-fact block; later denials emit a condensed single-line
  message that carries the denial ordinal, so consecutive denials are
  structurally different and never textually identical.
- True retries of the same target remain allowed without re-prompting
  (unchanged). Destructive-Bash and routine-Bash gates are unchanged,
  as are the ECC_GATEGUARD=off / ECC_DISABLED_HOOKS escape hatches.

Eight new tests cover budget counting, condensed format, ordinal
advancement, retry pass-through, env tuning, malformed state, MultiEdit
dampening, and destructive-gate exemption.

* fix(hooks): keep security hooks able to block on oversized stdin (#2222)

Refine the truncation fail-open: instead of skipping the hook entirely,
the runner now suppresses only its own raw-echo when stdin was
truncated. The hook still executes and receives the truncated flag
(run() context / ECC_HOOK_INPUT_TRUNCATED), so config-protection keeps
blocking truncated protected-config payloads (its test requires exit 2)
while pass-through hooks fail open with empty stdout as before.

* style: apply repo formatter to touched hook files
This commit is contained in:
Affaan Mustafa
2026-06-11 00:31:33 -04:00
committed by GitHub
parent 3bdb4a5e12
commit 6319c7d309
14 changed files with 846 additions and 151 deletions
Download Patch File Download Diff File Expand all files Collapse all files

149
tests/scripts/codex-hooks.test.js
View File

@@ -261,7 +261,7 @@ if (
else failed++;
if (
test('merge-mcp-config dry-run appends all recommended servers without mutating target', () => {
test('merge-mcp-config dry-run appends the current default set without mutating target', () => {
const tempDir = createTempDir('mcp-merge-dry-run-');
const configPath = path.join(tempDir, 'config.toml');
const original = '';
@@ -272,9 +272,12 @@ if (
assert.strictEqual(result.status, 0, `${result.stdout}\n${result.stderr}`);
assert.match(result.stdout, /Package manager: npm \(exec: npx\)/);
assert.match(result.stdout, /\[add\] mcp_servers\.supabase/);
assert.match(result.stdout, /\[mcp_servers\.github\]/);
assert.match(result.stdout, /\[add\] mcp_servers\.chrome-devtools/);
assert.match(result.stdout, /\[mcp_servers\.chrome-devtools\]/);
assert.match(result.stdout, /Dry run/);
// Retired defaults (June 2026 connector policy) must not be emitted.
assert.doesNotMatch(result.stdout, /mcp_servers\.(supabase|playwright|context7|exa|github|memory|sequential-thinking)\b/);
assert.doesNotMatch(result.stdout, /url = /);
assert.strictEqual(fs.readFileSync(configPath, 'utf8'), original);
} finally {
cleanup(tempDir);
@@ -296,14 +299,17 @@ if (
const merged = fs.readFileSync(configPath, 'utf8');
const parsed = TOML.parse(merged);
assert.strictEqual(parsed.mcp_servers.exa.url, 'https://mcp.exa.ai/mcp');
assert.strictEqual(parsed.mcp_servers.github.command, 'bash');
assert.deepStrictEqual(parsed.mcp_servers.memory.args, ['@modelcontextprotocol/server-memory']);
assert.strictEqual(parsed.mcp_servers.supabase.tool_timeout_sec, 120);
assert.strictEqual(parsed.mcp_servers['chrome-devtools'].command, 'npx');
assert.deepStrictEqual(parsed.mcp_servers['chrome-devtools'].args, ['chrome-devtools-mcp@latest']);
assert.strictEqual(parsed.mcp_servers['chrome-devtools'].startup_timeout_sec, 30);
// No retired server may be (re-)emitted — exa's url form broke Codex (#2224).
assert.strictEqual(parsed.mcp_servers.exa, undefined);
assert.strictEqual(parsed.mcp_servers.github, undefined);
assert.strictEqual(parsed.mcp_servers.supabase, undefined);
const second = runNode(mergeMcpConfigScript, [configPath], deterministicPackageEnv);
assert.strictEqual(second.status, 0, `${second.stdout}\n${second.stderr}`);
assert.match(second.stdout, /\[ok\] mcp_servers\.github/);
assert.match(second.stdout, /\[ok\] mcp_servers\.chrome-devtools/);
assert.match(second.stdout, /All ECC MCP servers already present/);
assert.strictEqual(fs.readFileSync(configPath, 'utf8'), merged);
} finally {
@@ -315,24 +321,88 @@ if (
else failed++;
if (
test('merge-mcp-config update dry-run reports canonical and legacy section refreshes', () => {
test('merge-mcp-config repairs the invalid exa url entry from earlier ECC versions (#2224)', () => {
const tempDir = createTempDir('mcp-merge-exa-repair-');
const configPath = path.join(tempDir, 'config.toml');
const original = [
'[mcp_servers.github]',
'command = "npx"',
'args = ["-y", "@modelcontextprotocol/server-github"]',
'',
'[mcp_servers.exa]',
'url = "https://mcp.exa.ai/mcp"',
'',
].join('\n');
try {
fs.writeFileSync(configPath, original);
const result = runNode(mergeMcpConfigScript, [configPath], deterministicPackageEnv);
assert.strictEqual(result.status, 0, `${result.stdout}\n${result.stderr}`);
assert.match(result.stdout, /\[repair\] mcp_servers\.exa/);
const updated = fs.readFileSync(configPath, 'utf8');
const parsed = TOML.parse(updated);
assert.strictEqual(parsed.mcp_servers.exa, undefined, 'invalid exa url entry must be removed');
assert.doesNotMatch(updated, /url = "https:\/\/mcp\.exa\.ai\/mcp"/);
// User-managed servers are untouched; current default is added.
assert.strictEqual(parsed.mcp_servers.github.command, 'npx');
assert.strictEqual(parsed.mcp_servers['chrome-devtools'].command, 'npx');
// Re-running must not re-introduce the invalid entry.
const second = runNode(mergeMcpConfigScript, [configPath], deterministicPackageEnv);
assert.strictEqual(second.status, 0, `${second.stdout}\n${second.stderr}`);
assert.doesNotMatch(fs.readFileSync(configPath, 'utf8'), /mcp_servers\.exa/);
} finally {
cleanup(tempDir);
}
})
)
passed++;
else failed++;
if (
test('merge-mcp-config leaves a user-managed stdio exa entry untouched', () => {
const tempDir = createTempDir('mcp-merge-exa-stdio-');
const configPath = path.join(tempDir, 'config.toml');
const original = [
'[mcp_servers.exa]',
'command = "npx"',
'args = ["-y", "mcp-remote", "https://mcp.exa.ai/mcp"]',
'startup_timeout_sec = 30',
'',
].join('\n');
try {
fs.writeFileSync(configPath, original);
const result = runNode(mergeMcpConfigScript, [configPath], deterministicPackageEnv);
assert.strictEqual(result.status, 0, `${result.stdout}\n${result.stderr}`);
assert.doesNotMatch(result.stdout, /\[repair\]/);
const parsed = TOML.parse(fs.readFileSync(configPath, 'utf8'));
assert.strictEqual(parsed.mcp_servers.exa.command, 'npx');
assert.deepStrictEqual(parsed.mcp_servers.exa.args, ['-y', 'mcp-remote', 'https://mcp.exa.ai/mcp']);
} finally {
cleanup(tempDir);
}
})
)
passed++;
else failed++;
if (
test('merge-mcp-config update dry-run refreshes managed sections and leaves user servers alone', () => {
const tempDir = createTempDir('mcp-merge-update-dry-run-');
const configPath = path.join(tempDir, 'config.toml');
const original = [
'[mcp_servers.chrome-devtools]',
'command = "custom"',
'args = ["old"]',
'',
'[mcp_servers.context7]',
'command = "custom"',
'args = ["old"]',
'',
'[mcp_servers.context7-mcp]',
'command = "npx"',
'args = ["legacy"]',
'',
'[mcp_servers.supabase]',
'command = "custom"',
'args = ["old"]',
'',
'[mcp_servers.supabase.env]',
'SUPABASE_ACCESS_TOKEN = "token"',
'args = ["-y", "@upstash/context7-mcp@latest"]',
'',
].join('\n');
@@ -341,11 +411,10 @@ if (
const result = runNode(mergeMcpConfigScript, [configPath, '--update-mcp', '--dry-run'], deterministicPackageEnv);
assert.strictEqual(result.status, 0, `${result.stdout}\n${result.stderr}`);
assert.match(result.stdout, /\[remove\] mcp_servers\.context7/);
assert.match(result.stdout, /\[remove\] mcp_servers\.context7-mcp/);
assert.match(result.stdout, /\[remove\] mcp_servers\.supabase/);
assert.match(result.stdout, /\[mcp_servers\.supabase\]/);
assert.match(result.stdout, /\[mcp_servers\.context7\]/);
assert.match(result.stdout, /\[remove\] mcp_servers\.chrome-devtools/);
assert.match(result.stdout, /\[mcp_servers\.chrome-devtools\]/);
// Retired servers are no longer ECC-managed: never removed or re-added.
assert.doesNotMatch(result.stdout, /\[remove\] mcp_servers\.context7/);
assert.strictEqual(fs.readFileSync(configPath, 'utf8'), original);
} finally {
cleanup(tempDir);
@@ -356,38 +425,31 @@ if (
else failed++;
if (
test('merge-mcp-config removes disabled legacy servers without appending replacements', () => {
test('merge-mcp-config removes disabled servers without appending replacements', () => {
const tempDir = createTempDir('mcp-merge-disabled-');
const configPath = path.join(tempDir, 'config.toml');
const original = [
'[mcp_servers.context7-mcp]',
'[mcp_servers.chrome-devtools]',
'command = "npx"',
'args = ["legacy"]',
'',
'[mcp_servers.exa]',
'url = "https://mcp.exa.ai/mcp"',
'args = ["chrome-devtools-mcp@latest"]',
'',
].join('\n');
const allServersDisabled = 'supabase,playwright,context7,exa,github,memory,sequential-thinking';
try {
fs.writeFileSync(configPath, original);
const result = runNode(mergeMcpConfigScript, [configPath], {
...deterministicPackageEnv,
ECC_DISABLED_MCPS: allServersDisabled,
ECC_DISABLED_MCPS: 'chrome-devtools',
});
assert.strictEqual(result.status, 0, `${result.stdout}\n${result.stderr}`);
assert.match(result.stdout, /Disabled via ECC_DISABLED_MCPS/);
assert.match(result.stdout, /\[skip\] mcp_servers\.context7 \(disabled\)/);
assert.match(result.stdout, /\[skip\] mcp_servers\.exa \(disabled\)/);
assert.match(result.stdout, /\[update\] mcp_servers\.context7-mcp \(disabled\)/);
assert.match(result.stdout, /\[update\] mcp_servers\.exa \(disabled\)/);
assert.match(result.stdout, /Done\. Removed 2 disabled server\(s\)\./);
assert.match(result.stdout, /\[skip\] mcp_servers\.chrome-devtools \(disabled\)/);
assert.match(result.stdout, /\[update\] mcp_servers\.chrome-devtools \(disabled\)/);
assert.match(result.stdout, /Done\. Removed 1 server section\(s\)\./);
const updated = fs.readFileSync(configPath, 'utf8');
assert.doesNotMatch(updated, /context7-mcp/);
assert.doesNotMatch(updated, /mcp_servers\.exa/);
assert.doesNotMatch(updated, /chrome-devtools/);
} finally {
cleanup(tempDir);
}
@@ -454,7 +516,10 @@ if (
assert.strictEqual(parsedConfig.agents.explorer.config_file, 'agents/explorer.toml');
assert.strictEqual(parsedConfig.agents.reviewer.config_file, 'agents/reviewer.toml');
assert.strictEqual(parsedConfig.agents.docs_researcher.config_file, 'agents/docs-researcher.toml');
assert.ok(parsedConfig.mcp_servers.exa);
// Current default connector is added; retired servers are not emitted,
// and pre-existing user-managed entries are preserved untouched.
assert.ok(parsedConfig.mcp_servers['chrome-devtools']);
assert.strictEqual(parsedConfig.mcp_servers.exa, undefined);
assert.ok(parsedConfig.mcp_servers.github);
assert.ok(parsedConfig.mcp_servers.memory);
assert.ok(parsedConfig.mcp_servers['sequential-thinking']);