sql-harness 0.2.4__tar.gz → 0.3.2__tar.gz

sql_harness-0.3.2/AGENTS.md

@@ -0,0 +1,19 @@
+# AGENTS.md — pointer
+
+> All contributor documentation lives in `docs/AGENTS.md` (single docs directory).
+> This pointer file exists only because agent frameworks auto-load `AGENTS.md`
+> by filename at the repo root — read the real content at `docs/AGENTS.md`.
+
+# Code priorities
+
+- Clarity
+- Precision
+- Low verbosity
+- Versatility
+
+# One-line summary
+
+sql-harness is a single-process SQL + SSH CLI for agents: plaintext
+connections in one TOML file, heredoc mode with auto-injected helpers.
+Core code in `src/sql_harness/`, agent-editable workspace under
+`agent-workspace/`. Full module list + contributing rules: `docs/AGENTS.md`.

{sql_harness-0.2.4 → sql_harness-0.3.2}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.5
 Name: sql-harness
-Version: 0.2.4
+Version: 0.3.2
 Summary: Single-process SQL + SSH CLI for LLM agents. PostgreSQL/MySQL/SSH via SQLAlchemy + paramiko. Plaintext credentials in one TOML file. Helpers auto-injected into the heredoc namespace.
 Project-URL: Source, https://github.com/zhaoliuxue/much_bigpy/tree/master/lab/sql_harness
 Project-URL: Issues, https://github.com/zhaoliuxue/much_bigpy/issues
@@ -25,7 +25,7 @@ Description-Content-Type: text/markdown
 
 A thin, single-process SQL CLI for LLM agents. Mirrors [browser-harness](https://github.com/browser-use/browser-harness)'s structure but targets relational databases (Postgres, MySQL, Redis-soon).
 
-Connection file format, env vars, and driver list: see `install.md`.
+Connection file format, env vars, and driver list: see `docs/install.md`.
 
 ## Quickstart
 
@@ -53,7 +53,7 @@ PY
 
 ## Architecture (~1k lines across 8 core files)
 
-- `install.md` — first-time install + first connection
+- `docs/install.md` — first-time install + first connection
 - `SKILL.md` — day-to-day usage
 - `lab/sql_harness/src/sql_harness/` — protected core package
 - `${XDG_CONFIG_HOME:-~/.config}/sql-harness/connections.toml` — plaintext credentials in ONE place
@@ -77,7 +77,7 @@ PY
 
 ## Contributing
 
-PRs and improvements welcome. See `AGENTS.md` for code priorities.
+PRs and improvements welcome. See `docs/AGENTS.md` for code priorities.
 
 - **Skills are written by the harness, not by you.** When you figure out a non-obvious SQL flow (a weird schema, a slow query, a JSON column trick), file a skill in `agent-workspace/zones/<conn>/skills/<name>.md` for that DSN, or `agent-workspace/zones/meta/skills/<name>.md` if it generalizes. Future sessions will read it before re-discovering it.
 - Bug fixes, new drivers, helper additions all welcome.

sql_harness-0.3.2/SKILL.md

@@ -0,0 +1,195 @@
+---
+name: sql-harness
+description: "Always use sql-harness for SQL, SSH, and database operations — querying any DSN (Postgres/MySQL/SQLite/Redis), inspecting schemas, running EXPLAIN plans, indexing analysis, schema migrations, shell exec over SSH, SFTP upload/download, and any cross-DSN workflow. Reach for it whenever the task touches a database, a remote host, or schema work — not just for `query()` calls."
+---
+
+# sql-harness
+
+A single-process SQL + SSH CLI for agents. 5 drivers (`postgres`, `mysql`, `sqlite`, `redis`, `ssh`) share one heredoc namespace. Connections live in plaintext in one TOML file.
+
+For setup, read `install.md`. For stuck-point mechanics, see `interaction-skills/` (cross-DB) and `interaction-skills/postgres/` (PG depth).
+
+## Reach for sql-harness when...
+
+- Any DB task: query, schema, migration, EXPLAIN, indexing, slow-query, stats
+- "Run this on the prod box": SSH workspace (`ssh_exec` / `ssh_upload` / `ssh_run_script`)
+- "Deploy this compose stack to remote": SSH + `interaction-skills/ssh/docker-via-ssh.md`
+- "The DB is behind a bastion / can't reach directly": SSH port-forward / tunnel
+- "Why is this query slow?": `explain_analyze` + `slow_queries` + `index_usage_stats`
+- "Which indexes are dead?": `unused_indexes` + `seq_scan_heavy`
+- "Compare row counts prod vs staging": `use_workspace("prod")` → `query()` → `use_workspace("staging")` → `query()`
+- "Save this heredoc as a reusable script": `sql-harness save <name> -c <conn>`
+- Anything where you would otherwise reach for `psql`, `mysql`, `pgcli`, or hand-rolled `subprocess` over SSH
+
+## Usage
+
+```bash
+sql-harness --help
+sql-harness list
+sql-harness add <name> --driver postgres --url 'postgresql://user:pw@host/db'
+sql-harness add <name> --driver mysql   --url 'mysql+pymysql://user:pw@host/db'
+sql-harness add <name> --driver ssh     --url 'ssh://user@host:22?key=/path/id_ed25519'
+sql-harness test <name>
+
+sql-harness <<'PY'
+use_workspace("local_pg")
+print(query("SELECT version()"))
+print(list_tables())
+PY
+```
+
+- Helpers are pre-imported; call by name in heredoc mode.
+- **First call requires `use_workspace(name)`** — no implicit default.
+- Set `BH_SQL_ZONE_SKILLS=1` once so `use_workspace()` auto-surfaces that zone's per-DSN skill docs (see `interaction-skills/zone-skill-auto-surface.md`).
+- For task-specific helpers, drop them into `agent-workspace/zones/<conn>/helpers.py` and they merge into the heredoc namespace (per-zone, overriding the shared base).
+- Other subcommands: `init` (scaffold connections.toml + install preset zone skills), `doctor` (SELECT 1 on every connection), `edit`/`remove`/`show`, `workspace {list,use,close,show}`, `save`/`run`/`scripts`, `skill` (bare: emit packaged SKILL.md; `install`/`list`/`show`), `paths`, `open`, `stats`, `version`, `ssh {exec,upload,download,run-script,info}`.
+- Zone-scoped subcommands (`save`, `run`, `scripts`, `skill list/show`, `ssh`) need `-c <conn>` or `$BH_SQL_ACTIVE_CONNECTION`.
+- **Analytics (ON by default)**: every CLI call and every doc read (`apply_skill`, `skill`, `skill show`) is counted into `$BH_SQL_HOME/analytics.json`; view with `sql-harness stats`. Every executed command (CLI argv or heredoc body) plus failures is appended to `$BH_SQL_HOME/sql-harness.log` (NDJSON, one record per line). Disable both with `BH_SQL_ANALYTICS=0|off|false|no`. Extension point: `from sql_harness.analytics import on, CLI_INVOKED, DOC_READ, COMMAND_EXECUTED` to attach custom listeners (observer pattern).
+
+## Drivers & capabilities
+
+Five drivers, one heredoc namespace. URL scheme must match the backend: MySQL requires `mysql+pymysql://`; PostgreSQL accepts `postgres://` / `postgresql://` (normalized to `postgresql+psycopg://`); SSH accepts `ssh://`, `ssh+password://`, `ssh+key://`.
+
+### Postgres (`postgres://`)
+
+Full SQLAlchemy + psycopg3 + 8 PG-only introspection helpers. Reach for these whenever the task is "PG + performance" or "PG + schema":
+
+- `explain(sql)` — `EXPLAIN` plan (estimated)
+- `explain_analyze(sql, buffers=True)` — `EXPLAIN (ANALYZE, BUFFERS)` (real times)
+- `table_stats(name, schema=None)` — `pg_stat_user_tables` (seq vs idx scan counts, dead tuples)
+- `index_usage_stats(table=None, schema=None)` — `pg_stat_user_indexes` (idx scan counts per index)
+- `unused_indexes(...)` — indexes with zero idx scans (candidates for DROP)
+- `seq_scan_heavy(...)` — tables routinely full-scanned (candidates for new indexes)
+- `slow_queries(limit=10)` — top by `pg_stat_statements.mean_exec_time`
+- `missing_indexes_hint(...)` — heuristic: many seq scans + table size > threshold
+
+See `interaction-skills/postgres/` for the 8 depth docs (plan-reading, B-Tree, specialized indexes, table optimization, slow-queries-joins, jsonb, migrations, indexes-and-explain).
+
+### MySQL (`mysql+pymysql://`)
+
+Standard SQLAlchemy + PyMySQL. JSON syntax differs from PG — see `interaction-skills/mysql/json-columns.md`.
+
+### SQLite (`sqlite:///path.db`)
+
+Local dev / tests. All standard helpers work. Use `sql-harness add --driver sqlite --url 'sqlite:///<absolute-path>'` (SQLite needs absolute path).
+
+### Redis (`redis://host:port`)
+
+Stub driver — wired up but minimal helpers. Prefer `query()` won't work (Redis is not SQL). Skip if your task needs Redis; flag in your reply.
+
+### SSH (`ssh://user@host:port?key=...`)
+
+**Not a database** — a remote shell workspace. Reach for ssh whenever the task says "on the box", "the prod host", "deploy this", "upload that file", "run that command remotely":
+
+```python
+use_workspace("prod-app")
+r = ssh_exec("systemctl status myapp")           # exit_code + stdout/stderr
+ssh_upload("./deploy.sh", "/tmp/deploy.sh")       # SFTP
+ssh_download("/var/log/app.log", "./app.log")     # SFTP
+ssh_run_script("./scripts/deploy.sh")             # upload + execute
+print(ssh_info())                                 # user/host/port/SFTP available
+```
+
+CLI equivalent: `sql-harness ssh -c <conn> {exec|upload|download|run-script|info} <args>`.
+
+See `interaction-skills/ssh/` for: `ssh.md` (helpers + CLI + gotchas), `docker-via-ssh.md` (docker-compose deploy loop over SSH), `auth-and-tunnels.md` (auth scheme 选型 + 堡垒机 + 端口转发).
+
+## Generic SQL helpers (all DB drivers)
+
+| Helper | Returns | Use |
+|---|---|---|
+| `query(sql, params=None)` | `list[dict]` | SELECTs only |
+| `execute(sql, params=None)` | `dict` (`lastrowid`, `rowcount`) | INSERT/UPDATE/DELETE/DDL |
+| `list_tables(schema=None)` | `list[str]` | Schema recon — always start here |
+| `describe(table, schema=None)` | `list[dict]` | Columns + types — before writing SQL |
+| `table(name, schema=None, limit=None)` | `list[dict]` | Quick peek: `SELECT * FROM x LIMIT n` |
+| `with_transaction()` | context manager yielding `Connection` | ≥ 2 statements / DDL / atomicity |
+| `run_sql_file(path)` | `list[dict]` | Multi-statement `.sql`, one transaction |
+| `server_version()` | `str` | Round-trip liveness check |
+| `connection_info()` | `dict` | Current workspace metadata |
+| `use_workspace_info(name)` | `dict` | Zone info + skills + scripts (always surfaces) |
+
+## Interaction Skills
+
+Stuck-point mechanics — read on demand by filename. Top-level (cross-DB):
+
+- `interaction-skills/aggregations.md` (+ `aggregations-advanced.md`)
+- `interaction-skills/encoding-and-charset.md`
+- `interaction-skills/joins.md` (+ `joins-advanced.md`)
+- `interaction-skills/large-result-sets.md`
+- `interaction-skills/migrations.md`
+- `interaction-skills/pooling.md`
+- `interaction-skills/recursive-ctes.md` (+ `recursive-ctes-advanced.md`)
+- `interaction-skills/save-run-cycle.md`
+- `interaction-skills/schema-introspection.md`
+- `interaction-skills/timeouts-and-cancellation.md`
+- `interaction-skills/transactions.md`
+- `interaction-skills/window-functions.md` (+ `window-functions-advanced.md`)
+- `interaction-skills/zone-skill-auto-surface.md`
+
+PG-specific depth — `interaction-skills/postgres/`:
+
+- `interaction-skills/postgres/btree-indexes.md`
+- `interaction-skills/postgres/indexes-and-explain.md`
+- `interaction-skills/postgres/json-columns.md`
+- `interaction-skills/postgres/migrations.md`
+- `interaction-skills/postgres/plan-reading.md`
+- `interaction-skills/postgres/slow-queries-joins.md`
+- `interaction-skills/postgres/specialized-indexes.md`
+- `interaction-skills/postgres/table-optimization.md`
+
+MySQL-specific — `interaction-skills/mysql/`:
+
+- `interaction-skills/mysql/json-columns.md`
+
+SSH-specific — `interaction-skills/ssh/`:
+
+- `interaction-skills/ssh/auth-and-tunnels.md`
+- `interaction-skills/ssh/docker-via-ssh.md`
+- `interaction-skills/ssh/ssh.md`
+
+## Cross-DSN strategy skills
+
+Reached via `apply_skill("pool")` etc. from any active zone — `zones/meta/skills/` is the passive fallback layer.
+
+- `agent-workspace/zones/meta/skills/pool.md` — pool sizing, `pre_ping`, idle reuse.
+- `agent-workspace/zones/meta/skills/workspace.md` — workspace isolation, multi-DSN workflows.
+
+## What actually works (field-tested)
+
+- **Start with `list_tables` + `describe`, not `query`.** Before writing SQL on an unknown DB, run `list_tables(schema=None)` then `describe("table", schema=None)` to learn the columns. Skips the trial-and-error SELECT cycle.
+- **Set `BH_SQL_ZONE_SKILLS=1` once** so every `use_workspace()` surfaces that zone's per-DSN skill docs as a hint list (10 filenames max). Then `apply_skill(name)` reads the body when relevant.
+- **Default to `with_transaction()`** for any write with ≥ 2 statements (insert + update, drop + create, …). Atomic + automatic rollback on exception.
+- **For PG performance tasks, reach for the 8 PG-only helpers first** (`explain_analyze`, `slow_queries`, `unused_indexes`, `seq_scan_heavy`, `table_stats`, `index_usage_stats`, `missing_indexes_hint`). Don't hand-parse `pg_stat_*` SQL.
+- **For remote tasks, reach for SSH workspace first.** Don't reach for raw `subprocess` + ssh-key files in agent code — `ssh_exec("systemctl restart myapp")` is one line.
+- **For "insert returning the new id"**: psycopg supports `INSERT … RETURNING id`; MySQL has no `RETURNING` — use `execute(...)` and read `result.lastrowid`.
+- **Streaming large result sets**: `query()` loads every row into memory. For >10k rows, use `ws.engine.connect().execution_options(stream_results=True)` and iterate manually.
+- **Read-only "is this alive?" check**: `print(server_version())` — single round-trip, works on every DSN, no per-driver fudging.
+- **Auth wall**: redirected to login → stop and ask the user. Don't type credentials from `connection_info`/`describe`.
+- **Credentials in TOML**: use `${env:VAR}` indirection for prod secrets. The connections.toml loader expands these at load time.
+- **Cross-DB SQL**: prefer ANSI syntax (`'string'`, `LIMIT n OFFSET m`, `COALESCE`, `CURRENT_TIMESTAMP`). PG-specific: `::TYPE`, `JSONB`, `RETURNING`. MySQL-specific: `AUTO_INCREMENT`, backticks, `?` placeholders (sqlalchemy normalizes to `:name`).
+- **Save the workflow**: every working block → `sql-harness save <name> -c <conn>`. Tomorrow's session re-runs with one command. See `interaction-skills/save-run-cycle.md`.
+
+## Design Constraints
+
+- One connection = one workspace; never share engines across workspaces.
+- Connection pool defaults: `size=5, recycle=3600, pre_ping=True` (per-connection overrides win).
+- `with_transaction()` yields a `Connection`; use `conn.execute(text(...))` for raw control.
+- `query()` returns `list[dict]`; use `execute()` for INSERT/UPDATE/DELETE.
+- `list_tables()` and `describe()` are read-only schema introspection helpers.
+- SSH workspaces expose `ssh_exec` / `ssh_upload` / etc. Calling them on a non-SSH workspace raises `RuntimeError`.
+- PG-only helpers (`explain_analyze`, `table_stats`, ...) raise `NotImplementedError` on non-PG workspaces. Detect with `ws.driver.name == "postgres"`.
+- All engines live in `SqlHarness`'s in-memory registry. Single-process, no daemon, no remote backend.
+
+## Gotchas
+
+- Driver label in TOML must match the URL scheme (`postgres` ⇄ `postgresql://`, `mysql` ⇄ `mysql+pymysql://`, `ssh` ⇄ `ssh://` / `ssh+password://` / `ssh+key://`).
+- Passwords in TOML are plaintext; use `${env:VAR}` indirection for prod secrets.
+- `query()` is for SELECTs only. For INSERTs, use `execute()`.
+- For tables > 10k rows, use `engine.connect().execution_options(stream_results=True)` and iterate manually — `query()` loads everything into memory.
+- SSH SFTP subsystem may be disabled on the remote host. `ssh_upload/ssh_download` raise `RuntimeError("SFTP subsystem not available")`; fall back to `tar -czf - | ssh host tar -xzf -` via `ssh_exec`.
+- Default `ssh_exec` timeout is 30 s. Long-running commands (e.g. `pg_dump`, `tar czf`) need explicit `timeout=`.
+
+## Domain / table skills
+
+Per-table or per-schema skills: drop a markdown file into `agent-workspace/zones/<conn>/skills/<name>.md`, read with `apply_skill(name)`. Cross-DSN strategy goes to `zones/meta/skills/` instead. PG optimization depth (plan reading, indexes, `table_stats`, `slow_queries`, …) lives in `interaction-skills/postgres/`.

{sql_harness-0.2.4 → sql_harness-0.3.2/docs}/AGENTS.md

@@ -15,11 +15,12 @@ Core code lives in `lab/sql_harness/src/sql_harness/`:
 - `manager.py` — SqlHarness class: holds engines, workspaces, skills registry
 - `helpers.py` — heredoc helpers (auto-imported)
 - `agent_loader.py` — loads `$BH_SQL_AGENT_WORKSPACE/agent_helpers.py`
+- `analytics.py` — event bus (observer) + JSON stats + NDJSON execution log: CLI call counts, doc-read tracking, command/error history; `BH_SQL_ANALYTICS` (default ON), view with `sql-harness stats`
 - `run.py` — heredoc entry; delegates to CLI otherwise
-- `cli.py` — argparse subparser CLI (~10 subcommands)
+- `cli.py` — argparse subparser CLI (18 subcommands)
 
 `SKILL.md` tells agents how to use the harness.
-`install.md` tells agents how to install and configure it.
+`docs/install.md` tells agents how to install and configure it.
 
 An agent operating the harness only edits inside `agent-workspace/`:
 - `agent_helpers.py` — task-specific query helpers the agent adds
@@ -28,4 +29,4 @@ An agent operating the harness only edits inside `agent-workspace/`:
 # Contributing
 Consider what is really needed. Prefer the smallest diff that fixes the bug.
 
-To add a backend: implement `drivers/<name>.py` against the Driver protocol, register in `drivers/__init__.py:get_driver()`, add `tests/unit/test_<name>.py` (see install.md §Drivers).
+To add a backend: implement `drivers/<name>.py` against the Driver protocol, register in `drivers/__init__.py:get_driver()`, add `tests/unit/test_<name>.py` (see `docs/install.md` §Drivers).

{sql_harness-0.2.4 → sql_harness-0.3.2/docs}/README.md

@@ -2,7 +2,7 @@
 
 A thin, single-process SQL CLI for LLM agents. Mirrors [browser-harness](https://github.com/browser-use/browser-harness)'s structure but targets relational databases (Postgres, MySQL, Redis-soon).
 
-Connection file format, env vars, and driver list: see `install.md`.
+Connection file format, env vars, and driver list: see `docs/install.md`.
 
 ## Quickstart
 
@@ -30,7 +30,7 @@ PY
 
 ## Architecture (~1k lines across 8 core files)
 
-- `install.md` — first-time install + first connection
+- `docs/install.md` — first-time install + first connection
 - `SKILL.md` — day-to-day usage
 - `lab/sql_harness/src/sql_harness/` — protected core package
 - `${XDG_CONFIG_HOME:-~/.config}/sql-harness/connections.toml` — plaintext credentials in ONE place
@@ -54,7 +54,7 @@ PY
 
 ## Contributing
 
-PRs and improvements welcome. See `AGENTS.md` for code priorities.
+PRs and improvements welcome. See `docs/AGENTS.md` for code priorities.
 
 - **Skills are written by the harness, not by you.** When you figure out a non-obvious SQL flow (a weird schema, a slow query, a JSON column trick), file a skill in `agent-workspace/zones/<conn>/skills/<name>.md` for that DSN, or `agent-workspace/zones/meta/skills/<name>.md` if it generalizes. Future sessions will read it before re-discovering it.
 - Bug fixes, new drivers, helper additions all welcome.

{sql_harness-0.2.4 → sql_harness-0.3.2/docs}/install.md

@@ -88,10 +88,14 @@ pool_size = 10
 | Var | Purpose |
 |---|---|
 | `BH_SQL_HOME` | Override state root (`~/.config/sql-harness` by default) |
+| `BH_SQL_CONFIG_DIR` | Override config dir (defaults to `BH_SQL_HOME`) |
 | `BH_SQL_CONFIG_FILE` | Override connections.toml path |
 | `BH_SQL_AGENT_WORKSPACE` | Override agent-workspace directory |
 | `BH_SQL_RUNTIME_DIR` | Override runtime state dir |
 | `BH_SQL_TMP_DIR` | Override temp dir |
+| `BH_SQL_ACTIVE_CONNECTION` | Active DSN zone for zone-scoped CLI (`save`/`run`/`scripts`/`skill list/show`) when no `-c` flag |
+| `BH_SQL_ZONE_SKILLS=1` | `use_workspace()` auto-surfaces the zone's skills/scripts |
+| `BH_SQL_ANALYTICS` | Analysis mode (CLI call counts + doc-read tracking + execution log). **Default ON**; set `0`/`off`/`false`/`no` to disable. Data: `$BH_SQL_HOME/analytics.json` (view with `sql-harness stats`) + `$BH_SQL_HOME/sql-harness.log` (append-only NDJSON: every executed command and its outcome) |
 | `BH_PG_URL` | Used by integration tests (skipped if unset) |
 | `BH_MYSQL_URL` | Used by integration tests (skipped if unset) |
 
@@ -115,7 +119,7 @@ To add another SSH host: append a `[[connections]]` block with `driver = "ssh"`
 | `ModuleNotFoundError: No module named 'psycopg'` | Run `uv sync`                                                         |
 | `sql-harness: command not found`                 | Use `uv run sql-harness ...` or activate `.venv`                      |
 | `password authentication failed`                 | Check URL; use `${env:VAR}` for prod secrets                          |
-| `unknown driver: 'X'`                            | Only `postgres`, `mysql`, `redis`, `sqlite` are built-in             |
+| `unknown driver: 'X'`                            | Only `postgres`, `mysql`, `redis`, `sqlite`, `ssh`/`ssh+password`/`ssh+key` are built-in             |
 | Heredoc mode complains `no workspace is active`   | You forgot `use_workspace(name)`                                      |
 | Heredoc mode complains `name 'X' is not defined`  | You forgot the import — all helpers are pre-imported, just call them  |
 

sql_harness-0.3.2/docs/maintainer.md

@@ -0,0 +1,58 @@
+# sql-harness standalone project
+
+## Quick reference for the maintainer
+
+| Task | Command |
+|---|---|
+| Install (editable dev mode) | `uv tool install --python 3.12 -e .` |
+| Build wheel + sdist | `uv build` |
+| Publish to TestPyPI | `uv publish --repository testpypi --token $TEST_PYPI_TOKEN` |
+| Publish to PyPI | `uv publish --token $PYPI_TOKEN` |
+| Refresh an agent's registered skill copy | `uv run sql-harness skill install` (or `skill > ~/.claude/skills/sql-harness/SKILL.md`) |
+
+## Why the SKILL.md sync step exists
+
+The in-package skill docs are **generated at build time**, not kept as
+hand-synced copies. `pyproject.toml` `[tool.hatch.build.targets.wheel.force-include]`
+maps the repo docs into the wheel:
+
+- `SKILL.md` → `sql_harness/_skills/SKILL.md`
+- `interaction-skills` → `sql_harness/_skills/interaction-skills`
+- `agent-workspace/zones/meta/skills` → `sql_harness/_skills/zones-meta-skills`
+
+The root `SKILL.md` is the **single source of truth** (v0.3.0 consolidation —
+the old `skills/sql-harness/SKILL.md` and `src/sql_harness/SKILL.md` copies
+were removed). After editing it, just `uv build` again. `sql-harness skill`
+reads `sql_harness/_skills/SKILL.md` from the installed package and falls
+back to the repo-root `SKILL.md` in a source checkout.
+
+## Build outputs
+
+```bash
+$ uv build
+Building source distribution...
+Building wheel from source distribution...
+Successfully built dist\sql_harness-0.3.0.tar.gz
+Successfully built dist\sql_harness-0.3.0-py3-none-any.whl
+```
+
+The wheel contains the package modules + the `_skills/` doc bundle
+(SKILL.md + interaction-skills tree + zones-meta-skills).
+
+## Test before publish
+
+```bash
+# standalone venv
+uv sync
+uv run pytest tests/unit -q    # 80 tests must pass
+uv run sql-harness --help      # CLI works
+uv run sql-harness skill       # SKILL.md emit works
+```
+
+## Local PyPI dry-run (no upload)
+
+```bash
+uv publish --dry-run
+# → checks both dist files against upload.pypi.org/legacy/
+# → errors with "no OIDC token" (expected; pass --token to actually publish)
+```

{sql_harness-0.2.4 → sql_harness-0.3.2}/docs/standalone-install.md

@@ -162,6 +162,6 @@ monorepo access to `from sql_harness import ...` for testing — no breakage.
 
 ## See also
 
-- `lab/sql_harness/install.md` — first-time install + troubleshooting
-- `lab/sql_harness/AGENTS.md` — code priorities + extension recipe
-- `lab/sql_harness/README.md` — user-facing quickstart
+- `lab/sql_harness/docs/install.md` — first-time install + troubleshooting
+- `lab/sql_harness/docs/AGENTS.md` — code priorities + extension recipe
+- `lab/sql_harness/docs/README.md` — user-facing quickstart

sql_harness-0.3.2/interaction-skills/mysql/json-columns.md

@@ -0,0 +1,43 @@
+# JSON columns — MySQL (`JSON`)
+
+MySQL 5.7+ has a native `JSON` type. PyMySQL serializes Python `dict`/`list` to `JSON`.
+
+## Insert
+
+SQLAlchemy handles the conversion when passing a dict as a parameter:
+
+```python
+execute("INSERT INTO events (data) VALUES (:d)", {"d": {"event": "click"}})
+```
+
+## Query
+
+Use `JSON_EXTRACT(col, '$.path')` or the `->` / `->>` shorthand (MySQL 8.0+):
+
+```python
+rows = query("SELECT data->>'$.event' AS event FROM events WHERE JSON_EXTRACT(data, '$.event') = :e",
+             {"e": "click"})
+```
+
+Note `->>` in MySQL returns `TEXT` (unquoted). In PostgreSQL `->>` returns `text` of a jsonb path. The `'$.event'` path syntax is MySQL-specific.
+
+## Indexing
+
+MySQL 8.0+ supports functional indexes on JSON expressions:
+
+```sql
+CREATE INDEX idx_events_event ON events ((CAST(data->>'$.event' AS CHAR(64))));
+```
+
+Pre-8.0: index a generated column instead.
+
+## Gotchas
+
+- `JSON_EXTRACT` returns JSON; `->>` returns the unquoted scalar. For comparisons, match the operator.
+- `->>` shorthand requires MySQL 8.0+. On 5.7 use `JSON_UNQUOTE(JSON_EXTRACT(...))`.
+- MySQL `JSON` does NOT auto-compress (unlike PG `jsonb` TOAST). Large docs blow up row size.
+- Whitespace handling: MySQL normalizes on input but may reformat on output (PG normalizes the same way).
+
+## See also
+
+- `interaction-skills/postgres/json-columns.md` — PostgreSQL `jsonb` syntax (`@>`, `?`, no `$.path`)

sql_harness-0.3.2/interaction-skills/postgres/json-columns.md

@@ -0,0 +1,46 @@
+# JSON columns — PostgreSQL (`jsonb`)
+
+Use PostgreSQL's binary `jsonb` type (not `json` — `jsonb` is binary-comparable and indexable).
+
+## Insert
+
+`psycopg3` auto-serializes Python `dict`/`list` to `jsonb`:
+
+```python
+execute("INSERT INTO events (data) VALUES (:d)", {"d": {"event": "click", "ts": 1234}})
+```
+
+## Query
+
+Use `->` (returns `jsonb`), `->>` (returns `text`), and the containment operator `@>`:
+
+```python
+rows = query("SELECT data->>'event' AS event FROM events WHERE data @> :filter",
+             {"filter": {"event": "click"}})
+```
+
+## Indexing
+
+`jsonb` supports GIN indexes on the whole document (fast for `@>` / `?` / `?&` / `?|`):
+
+```sql
+CREATE INDEX idx_events_data ON events USING GIN (data);
+```
+
+For paths you hit often, a btree expression index:
+
+```sql
+CREATE INDEX idx_events_event ON events ((data->>'event'));
+```
+
+## Gotchas
+
+- Prefer `jsonb` over `json` — `json` is text-stored, no index support beyond expression indexes, whitespace-preserved.
+- Comparing `data = '{"event":"click"}'::jsonb` works (SQLAlchemy normalizes JSON whitespace).
+- `JSONB` aggregates (`jsonb_agg`, `jsonb_object_agg`) exist; for plain text JSON use `string_agg`.
+- GIN on `jsonb` is large — measure before adding to high-write tables.
+
+## See also
+
+- `interaction-skills/postgres/specialized-indexes.md` — GIN/GiST/BRIN tradeoffs
+- `interaction-skills/mysql/json-columns.md` — MySQL JSON syntax (different operators)

sql_harness-0.3.2/interaction-skills/postgres/migrations.md

@@ -0,0 +1,105 @@
+# PG Data Movement
+
+Move data in/out/around PostgreSQL: logical dump + restore, bulk COPY, and
+logical replication. (Schema migrations -- ALTER / CREATE under transaction
+-- live in `migrations.md`; this file is for moving data across databases or
+in/out of tables.)
+
+## Detection
+
+- **Move a whole database** to a new server -> `pg_dump` + `pg_restore`.
+- **Bulk-load** a big table (>1M rows) -> `COPY` (10-100x faster than `INSERT`).
+- **Live replication** of changes from one PG to another (zero-downtime
+  migration, read replica, aggregation) -> logical replication (publication
+  + subscription).
+- **One-shot export** for analysis / downstream pipeline -> `COPY` /
+  `psql \copy`.
+
+## Approach
+
+### 1. Logical dump + restore (server-side CLI)
+
+`pg_dump` and `pg_restore` live on the PG host, not in SQL. Run them via
+the SSH workspace that hosts your PG:
+
+```python
+use_workspace("2026aliyun")   # SSH workspace on the PG host
+
+ssh_exec("pg_dump -Fc -d sourcedb -f /tmp/sourcedb.dump")     # custom format, compressed
+ssh_exec("createdb targetdb")                                  # empty target
+ssh_exec("pg_restore -d targetdb --no-owner --jobs=4 /tmp/sourcedb.dump")
+```
+
+Flags worth knowing:
+
+- `-Fc` custom format (compressed, parallel-restoreable). `-Fp` plain SQL.
+- `-s` schema-only / `-a` data-only. `-t public.orders` one table.
+- `--no-owner` strips `OWNED BY` from objects (safer cross-cluster).
+
+Cross-version note: `pg_dump` from the **target's** major is safest.
+`pg_dump` 16 can read PG 15 / 14; restore with the target's own
+`pg_restore`. Going major-down (PG 16 dump -> PG 14) is **not supported**.
+
+### 2. Bulk COPY
+
+`COPY` bypasses per-row parse/plan/execute -- it's the right tool for
+anything >10k rows. `sql-harness`'s `execute` covers server-side file
+reads; client-streaming needs psql's `\copy` via SSH.
+
+```python
+use_workspace("2026aliyun")
+
+# (a) COPY TO -- export a query to a server-side CSV. psql \copy wraps
+#     COPY and lets a client feed/collect the stream.
+ssh_exec(r"""psql -U $PG_USER sourcedb -c "\copy (SELECT id,email FROM users WHERE created_at>'2025-01-01') TO '/tmp/users.csv' WITH CSV HEADER" """)
+ssh_download("/tmp/users.csv", "./users.csv")     # pull it local
+
+# (b) COPY FROM file -- server reads directly. Needs superuser OR
+#     `GRANT pg_read_server_files TO <role>` (PG 14+).
+execute("COPY users (id, email) FROM '/var/lib/pg/imports/users.csv' WITH CSV HEADER")
+
+# (c) COPY FROM STDIN -- client streams. `execute` can't do streaming;
+#     psql \copy from a *local* file is the practical path:
+ssh_exec(r"""psql -h $PG_HOST -U $PG_USER targetdb -c "\copy users (id,email) FROM '/local/users.csv' WITH CSV HEADER" """)
+```
+
+### 3. Logical replication (publication + subscription)
+
+For live, ongoing PG-to-PG replication of row changes. SQL-only on both
+sides; the only server config is `wal_level = logical` on the source.
+
+```python
+use_workspace("source_pg")
+# Source: declare what to publish. Default = all DML on the named tables.
+execute("CREATE PUBLICATION pub_users FOR TABLE users")
+# Add `WITH (publish = 'insert,update,delete,truncate')` to include TRUNCATEs.
+
+use_workspace("target_pg")
+# Target: subscribe. Initial copy runs, then changes stream.
+execute("""CREATE SUBSCRIPTION sub_users
+           CONNECTION 'host=src.db port=5432 user=repl password=... dbname=source'
+           PUBLICATION pub_users""")
+
+# Monitor catch-up from the TARGET:
+for r in query("SELECT subname, status, received_lsn, latest_end_lsn FROM pg_stat_subscription"):
+    print(r)      # status='streaming' + latest_end_lsn advancing = caught up
+```
+
+## Gotchas
+
+- **Logical replication does NOT replicate**: schema (DDL), sequence state
+  (manually sync with `pg_dump --section=pre-data` on the source +
+  `pg_restore --section=post-data` on target), or `TRUNCATE` unless the
+  publication includes `publish='truncate'`.
+- **Tables need a PK** (or explicit `REPLICA IDENTITY`) -- without it,
+  `UPDATE`/`DELETE` can't replicate.
+- **`COPY FROM` file path** is server-side (`SHOW data_directory`).
+  Permissions: superuser OR `pg_read_server_files` (PG 14+).
+- **`pg_dump` cross-major** only goes *up* (or equal). PG 14 dump -> PG 16
+  is fine; PG 16 dump -> PG 14 is rejected.
+- **`pg_restore --jobs=N`** uses N cores but each job = one transaction
+  per table -- mid-restore failures leave the DB half-loaded. For
+  restartability, restore one table at a time.
+- **Logical replication initial copy** puts write load on the source; for
+  very large DBs, stop writes briefly or use `ALTER SUBSCRIPTION ...
+  DISABLE` then re-enable after catching up.