dsh-headless-json

Structured, machine-readable CI output for DeepSeek Harness (dsh).
> 中文文档:README.zh.md
dsh's built-in headless mode prints the final assistant text and exits 0/1 — perfect for a quick smoke test, but thin for CI. dsh-headless-json is a profile bundle that turns any dsh session into a first-class CI artifact:
- a transaction-level JSON report (session metadata, every structured
event, outcome, statistics),
- an append-only NDJSON event stream you can
tailwhile the run is live, - a JUnit XML report for GitLab CI / Jenkins / Azure DevOps / etc.,
- semantic exit codes that distinguish success, failure, timeout,
blocking, abort and interruption,
- an artifact manifest of file paths referenced in the conversation,
- a privacy layer (text truncation, argument hiding, secret masking,
path relativization),
- and a dual entry point: dsh tools inside the runtime
(output_status, output_events, set_options) plus a standalone CLI for offline rendering and exit-code plumbing.
Everything is deterministic: given the same session log and the same options, the produced bytes are identical.
---
Contents
- [Features](#features)
- [How it works](#how-it-works)
- [Bundle layout](#bundle-layout)
- [Installation & integration](#installation--integration)
- [What a run produces](#what-a-run-produces)
- [dsh tools](#dsh-tools)
- [CLI reference](#cli-reference)
- [Configuration](#configuration)
- [Output formats](#output-formats)
- [JSON report](#json-report) - [Events](#events) - [NDJSON stream](#ndjson-stream) - [JUnit XML](#junit-xml)
- [Exit code semantics](#exit-code-semantics)
- [Privacy & redaction](#privacy--redaction)
- [Artifacts](#artifacts)
- [Determinism](#determinism)
- [Development](#development)
- [Limitations & compatibility](#limitations--compatibility)
---
Features
| Area | What you get | | --- | --- | | Session event streaming | Subscribes to the dsh session event firehose (session/created, session/event, session/flush, session/disposed) and derives one structured event per turn/step/tool call: type, model, latency, tokens, tool name, argument summary, result, error, status. | | JSON output | A complete transaction-level report: session metadata + event list + outcome/exit code + statistics. Configurable incremental NDJSON stream. | | JUnit XML output | Tool calls, steps and turns map to test cases, so any CI system displays a dsh run like a test run. | | Exit-code semantics | Stable categories (success / error / timeout / blocked / empty / aborted / interrupted) with documented default codes, all overridable. | | Artifact collection | File paths referenced in user/assistant/tool texts are collected into a manifest with existence/size when running on the same machine. | | Privacy switch | Tool output truncation, argument hiding, secret masking, cwd-relative paths — configurable at mount time and at runtime via set_options. | | Toolchain | Three dsh tools (output_status, output_events, set_options) plus a standalone CLI (dsh-headless-json render|exit). |
How it works
The bundle is a dsh profile bundle: an npm package whose manifest declares
"dsh": { "bundle": { "patch": "./cordis.patch.yml" } }The patch inserts one plugin (headless-json) into the loader tree. The plugin subscribes to the session service events, keeps one capture per session, streams NDJSON lines as events arrive, and writes the JSON/JUnit reports when the session is disposed (which happens during dsh's normal graceful teardown).
dsh session event stream
│ session/created · session/event · session/flush · session/disposed
▼
┌─────────────────── CaptureManager ───────────────────┐
│ per-session Capture: derive structured events │
│ (turn/step/tool latency, tokens, model, errors) │
│ + artifact scan + type distribution │
└────────────┬──────────────────────────┬─────────────┘
│ live NDJSON lines │ on session end
▼ ▼
events.ndjson report.json · junit.xml- Capture vs. output: events are derived once, in memory; redaction is
applied at output time, so changing options with set_options affects every subsequent write.
- No stdout interference: the plugin never writes to stdout, so it
composes cleanly with the official headless runner (which owns stdout).
- Error containment: every listener is wrapped; a malformed event can
never crash the tree or corrupt a report.
Bundle layout
headless-json/
├── package.json # dsh.bundle.patch manifest + bin
├── cordis.patch.yml # loader patch (inserts the plugin row)
├── lib/ # compiled ESM (tsc output)
├── bin/dsh-headless-json.js # CLI entry
├── src/ # TypeScript sources (repo only)
├── test/ # node:test suite (repo only)
├── examples/ # sample session log + generated outputs
├── README.md / README.zh.md
└── LICENSE # MITThe repository tree is shown above; the published npm package ships lib, bin, cordis.patch.yml, examples, both READMEs and LICENSE.
Installation & integration
Requirements: Node.js ≥ 18, a dsh installation that composes the sessions service (any base/headless profile does).
1. Build (only when developing from source)
npm install
npm run build # or: npm test (builds + runs the suite)2. Add the bundle to a dsh profile
Using the CLI (forwards to pnpm):
dsh plugin --profile headless add /path/to/headless-jsonor straight from this repository:
dsh plugin --profile headless add github:JohnXu22786/headless-jsonor manually: add dsh-headless-json to the profile's package.json dependencies and to the ordered dsh.profile.bundles list, then pnpm install in the profile directory. dsh reconciles installed bundles against dsh.profile.bundles on the next run.
3. Run
dsh --profile headless "run the test suite"After the run completes you get, in dsh-output/ (relative to the working directory):
dsh-output/report.json # transaction-level JSON report
dsh-output/junit.xml # JUnit XML report
dsh-output/events.ndjson # only when output.ndjson = true4. Wire CI
The semantic exit code lives in the report; the CLI turns it into a process exit:
dsh --profile headless "run the test suite"
code=$(dsh-headless-json exit dsh-output/report.json)
exit $codeor pipe a JUnit report straight into your CI collector:
dsh-headless-json render dsh-output/report.json --format junit --out junit.xmlWhat a run produces
Per session, the plugin writes:
| File | When | Content |
|---|---|---|
report.json | session end | Full report (see [JSON report](#json-report)) |
junit.xml | session end | JUnit XML (see [JUnit XML](#junit-xml)) |
events.ndjson | live | One JSON object per event, appended in real time, plus a final session_end line |
File names honor a {session} placeholder (e.g. "json_file": "report-{session}.json"); without it, a second session in the same process gets a -<id> suffix before the extension. A session that recorded zero events writes nothing when output.write_empty is false.
dsh tools
The bundle registers three tools (via ctx.tools.register) visible to every agent. They are resolved against the calling agent's session.
| Tool | Arguments | Returns |
|---|---|---|
output_status | none | Live capture status: event/turn/tool counts, pending tool calls, redaction settings, streaming state. |
output_events | types? (array of kinds), since? (seq), limit? (default 200) | The requested structured events, redacted exactly like the files. |
set_options | partial output / redact / artifacts / events / exit | The new effective options after validation. |
Example conversation:
set_options({ "redact": { "args": "hide" } })
→ { "applied": true, "redact": { "args": "hide", ... } }CLI reference
dsh-headless-json render <input> [options] Render a report/event file
dsh-headless-json exit <input> [options] Print the semantic exit code
dsh-headless-json --version | --help<input> may be:
- a
report.jsonproduced by the plugin, - an
events.ndjsonstream produced by the plugin, - a raw dsh session
.jsonllog (header line +session/*event lines), - a JSON array of raw session events.
| Option | Meaning | | --- | --- | | --format json\|junit\|ndjson | Output format (default json). | | --out <file> | Write to a file instead of stdout. | | --pretty | Pretty-print the JSON report. | | --set <key=value> | Override an option, e.g. --set redact.text_length=1200. Repeatable. | | --text-length <n> | Shortcut for --set redact.text_length=<n>. | | --arg-length <n> | Shortcut for --set redact.arg_length=<n>. | | --args full\|truncate\|hide | Shortcut for --set redact.args=<mode>. | | --paths relative\|absolute | Shortcut for --set redact.paths=<mode>. | | --no-secrets | Disable secret masking. | | --max-events <n> | Shortcut for --set events.max_events=<n>. | | --cwd <dir> | Base directory used for path relativization. | | --include-log-only | Surface log-only event types as other. |
Notes:
exitprints the code and exits the process with it — the CI glue.- Redaction
--setoverrides re-render raw session logs only. An
events.ndjson stream and a report.json are already redacted (redaction was applied when each line/file was written at capture time) and are re-rendered verbatim.
Configuration
All options are snake_case, configurable through the bundle's patch config block, the set_options tool, or CLI --set. Unknown keys are tolerated in configuration files (forward-compatible deployments); set_options and CLI --set reject unknown keys, including unknown nested keys.
# cordis.patch.yml (profile overlay example)
- id: headless-json
config:
output:
dir: dsh-output
json: true
junit: true
ndjson: true
redact:
text_length: 4000
args: truncate
paths: relative
secrets: true| Key | Default | Description |
|---|---|---|
output.dir | dsh-output | Report directory (relative to cwd). |
output.json | true | Write report.json. |
output.junit | true | Write junit.xml. |
output.ndjson | false | Stream events.ndjson live. |
output.json_file | report.json | JSON file name ({session} placeholder supported). |
output.junit_file | junit.xml | JUnit file name. |
output.ndjson_file | events.ndjson | NDJSON file name. |
output.write_empty | true | Write reports for sessions with zero events. |
output.pretty | false | Pretty-print the JSON report. |
redact.text_length | 4000 | Max chars of any text value in outputs (0 = unlimited). |
redact.arg_length | 500 | Max chars of the tool-argument summary. |
redact.args | truncate | full (parsed object), truncate (summary), hide ([hidden]). |
redact.paths | relative | relative (against session cwd) or absolute. |
redact.secrets | true | Mask secret-shaped strings. |
redact.secret_patterns | [] | Extra regex source strings, applied in addition to the built-ins. |
artifacts.collect | true | Scan texts for file-path references. |
artifacts.check_exists | true | stat matched paths (only meaningful on the run machine). |
artifacts.max_entries | 500 | Maximum unique artifact entries. |
artifacts.pattern_extras | [] | Extra path regex source strings. |
events.max_events | 1000 | Max events in the report's event list (0 = unlimited). |
events.trim | balanced | head keeps the first N; balanced keeps head+tail. |
events.include_log_only | false | Surface log-only types (request/header, session/end-seed, …) as other events. |
exit.* | see table | Per-category exit codes (see [Exit code semantics](#exit-code-semantics)). |
capture.text_cap | 100000 | Hard cap on the chars of any single stored text. |
Output formats
JSON report
{
"schema_version": 1,
"plugin": { "name": "dsh-headless-json", "version": "0.1.0" },
"generated_at": 1753000000000, // session end time (derived, deterministic)
"session": {
"id": "...", "cwd": ".", "cwd_name": "repo",
"created_at": ..., "started_at": ..., "ended_at": ...,
"event_count": 42,
"parent_session": null, "agent_preset": "minimal", "delegation_depth": null
},
"outcome": {
"status": "success", // stable category
"exit_code": 0, // semantic exit code
"reason": "completed", // raw turn/end reason kind
"complete": true,
"undelivered_tool_calls": [],
"error": null // {code, message, status?} on error
},
"stats": {
"duration_ms": 4123.5,
"turns": 1, "steps": 2,
"assistant_messages": 3, "user_messages": 1,
"tool_calls": 2, "tool_errors": 0, "undelivered_tool_calls": 0,
"chunk_count": 14,
"tokens": { "input": 900, "output": 300, "cache_read": 0, "cache_write": 0, "reasoning": 0 },
"events_by_type": { "assistant/message": 3, "tool/call": 2, ... }, // raw dsh types
"by_tool": { "bash": { "calls": 2, "errors": 0, "latency_ms_total": 800, "latency_ms_max": 500 } }
},
"events": [ /* see below */ ],
"events_truncated": null, // {kept,total,dropped} when trimmed
"artifacts": [ { "path": "src/options.ts", "kind": "file", "size": 4281, "references": 3 } ]
}Events
Every event carries seq (session-log sequence, the stable sort key), time (unix ms) and kind:
| kind | extra fields |
|---|---|
turn_start | turn |
turn_end | turn, reason, error (code/message/status), cause (abort kind), latency_ms, complete |
step_start / step_end | turn, step, (latency_ms) |
user_message | turn, step, text, reasoning, blocks |
assistant_message | turn, step, provider, model, latency_ms, usage, text, reasoning, blocks, stream (chunk count/times) |
tool_call | turn, step, call_id, tool, args_mode, args, args_summary, latency_ms, undelivered |
tool_result | turn, step, call_id, tool, status (success/error), latency_ms, text, blocks, error |
todo_write | count, todos |
request_context | provider, model, context_window |
other | type (only with events.include_log_only) |
Tool calls and results are correlated by call_id; latencies are computed from the session log timestamps. A tool_call whose tool/result never arrived by session end is flagged undelivered: true and listed in outcome.undelivered_tool_calls.
NDJSON stream
One JSON object per line, in arrival order, written immediately: the same structured events plus a final summary line:
{"kind":"session_end","session_id":"...","generated_at":...,
"outcome":{"status":"success","exit_code":0,"reason":"completed"},
"stats":{"duration_ms":4123,"turns":1,"steps":2,"tool_calls":2,"tool_errors":0}}The file is fsynced on every session/flush checkpoint, so consumers that read storage after whenIdle() see a durable stream. A tool_call line is streamed before its result arrives, so the latency_ms / undelivered fields on those lines are provisional; the correlating tool_result line (or the final report) carries the authoritative values.
JUnit XML
| dsh concept | JUnit mapping |
|---|---|
| the whole session | <testsuites name="dsh-headless-json"> → one <testsuite> with session properties (id, cwd, status, exit_code, reason, plugin version) |
| overall outcome | <testcase name="run"> — passes only when outcome.status === "success", otherwise <error>/<failure> |
| each turn | <testcase name="turn-N"> — completed passes; error/max-tokens fail; blocked/aborted/interrupted skip |
| each step | <testcase name="step-N"> with its duration; skipped when it never closed |
| each tool call | <testcase name="tool:NAME"> with its duration and the result text in <system-out>; failed calls produce <failure type="dsh:tool"> |
Timing is in decimal seconds, timestamps are ISO-8601 UTC, and all text is XML-1.0-escaped (control characters replaced).
Exit code semantics
The final turn/end reason maps to a stable category and a documented exit code; every value is overridable via exit.*:
| Category | Final turn/end reason | Default code |
|---|---|---|
success | completed | 0 |
error | error | 1 |
timeout | max-tokens | 2 |
blocked | blocked | 3 |
empty | (no events at all) | 4 |
aborted | aborted | 130 |
interrupted | interrupted | 130 |
Sessions with events but no closing turn/end are reported as error/incomplete. The 130 code intentionally follows the SIGINT convention for user-initiated stops.
Privacy & redaction
Defaults are conservative but usable. Everything applies at output time.
- Text truncation — every text/reasoning value is capped at
redact.text_length with a deterministic …[+N more chars] marker.
- Argument handling —
redact.args:
- full: the parsed argument object, string values masked/truncated; - truncate (default): a masked summary capped at redact.arg_length; - hide: the literal string [hidden].
- Secret masking (
redact.secrets) — built-ins coversk-…keys,
Bearer … headers, PEM private keys, GitHub/Google tokens, JWTs, long hex values (keeps a short prefix for e.g. commit SHAs) and token-like strings. Turn/end error messages are treated as free text and masked too. Add your own with redact.secret_patterns (regex source strings, validated at configuration time).
- Path relativization —
redact.paths: relative(default) strips the
session cwd from artifact paths and renders session.cwd as .; absolute keeps full paths.
What this means in practice: the default output contains no absolute workspace paths, no raw secret strings, and no unbounded tool output.
Artifacts
When artifacts.collect is on, every user/assistant/tool text is scanned for path-like candidates (POSIX/Windows absolute paths and separator-containing relative paths; URLs and emails are excluded). Each unique candidate becomes an artifact entry with a reference count; when artifacts.check_exists is on (the default) and the run happens on the same machine, kind (file/dir/missing) and size are filled in from the filesystem. The manifest is bounded by artifacts.max_entries.
Determinism
Given the same session events and the same effective options:
- the JSON report is byte-identical across runs — object key order is
fixed by construction, events are sorted by seq, map-like fields are serialized with sorted keys, and numbers are rounded to 3 decimals;
- JUnit XML and NDJSON lines are produced from the same serialization core,
so all three formats always agree.
generated_at (and the JUnit timestamp) are derived from the session's own timeline — the last event time — so they never depend on a wall clock.
Development
npm install
npm run build # tsc -> lib/
npm test # build + run the full suite (node:test)
npm run test:only # run tests against the current build
npm run typecheck # tsc --noEmitThe test suite covers e
…