Implement the kata behind an answer.sh at the project root: called as
sh answer.sh <string> <count> it prints the string with every character
repeated count times, and nothing else. The string is 1-14 characters of
[A-Za-z0-9_.-] starting with an alphanumeric; the count is 1-5.

Also write an AGENTS.md (or README.md) describing your stack (language,
tooling) and carrying the commands kept in session memory: a
run: line declaring the command that runs your solution, e.g.
run: sh answer.sh or run: python3 g.py. The platform re-reads those
files after every completed task and the following tasks invoke exactly
what the line declares - so you may switch language at any rung. It has
not been read yet during this task, which is why sh answer.sh is the
starting contract here. Declare a test: line there too - the command
that runs your tests (e.g. test: sh test.sh); it is captured into
session memory the same way. Size does not matter yet - the golf starts on the
next task.

Shrink the solution to at most 130 bytes while keeping it correct. Probes run whatever command your AGENTS.md/README.md "run" line declares - the platform re-reads it after every completed task - so you may rewrite in any language at any rung and the checks follow. One caveat - the server re-runs your claimed checks against the committed solution to verify the score, and its sandbox provides sh, node and python3, so behind any other interpreter your solution goes unverified rather than confirmed (it is never penalised for that). The size is the combined byte count of every file in the project except .git/, .ololo/, AGENTS.md, README.md and .gitignore, plus the declared run command itself - helper files and a leftover answer.sh count too, and moving code into the run line, or into a doc file the run line invokes, does not make it free (both are counted). Correctness is re-probed on every check with a freshly sampled string and count.

Shrink the solution to at most 110 bytes while keeping it correct. Probes run whatever command your AGENTS.md/README.md "run" line declares - the platform re-reads it after every completed task - so you may rewrite in any language at any rung and the checks follow. One caveat - the server re-runs your claimed checks against the committed solution to verify the score, and its sandbox provides sh, node and python3, so behind any other interpreter your solution goes unverified rather than confirmed (it is never penalised for that). The size is the combined byte count of every file in the project except .git/, .ololo/, AGENTS.md, README.md and .gitignore, plus the declared run command itself - helper files and a leftover answer.sh count too, and moving code into the run line, or into a doc file the run line invokes, does not make it free (both are counted). Correctness is re-probed on every check with a freshly sampled string and count.

Shrink the solution to at most 95 bytes while keeping it correct. Probes run whatever command your AGENTS.md/README.md "run" line declares - the platform re-reads it after every completed task - so you may rewrite in any language at any rung and the checks follow. One caveat - the server re-runs your claimed checks against the committed solution to verify the score, and its sandbox provides sh, node and python3, so behind any other interpreter your solution goes unverified rather than confirmed (it is never penalised for that). The size is the combined byte count of every file in the project except .git/, .ololo/, AGENTS.md, README.md and .gitignore, plus the declared run command itself - helper files and a leftover answer.sh count too, and moving code into the run line, or into a doc file the run line invokes, does not make it free (both are counted). Correctness is re-probed on every check with a freshly sampled string and count.

Shrink the solution to at most 80 bytes while keeping it correct. Probes run whatever command your AGENTS.md/README.md "run" line declares - the platform re-reads it after every completed task - so you may rewrite in any language at any rung and the checks follow. One caveat - the server re-runs your claimed checks against the committed solution to verify the score, and its sandbox provides sh, node and python3, so behind any other interpreter your solution goes unverified rather than confirmed (it is never penalised for that). The size is the combined byte count of every file in the project except .git/, .ololo/, AGENTS.md, README.md and .gitignore, plus the declared run command itself - helper files and a leftover answer.sh count too, and moving code into the run line, or into a doc file the run line invokes, does not make it free (both are counted). Correctness is re-probed on every check with a freshly sampled string and count.

Shrink the solution to at most 70 bytes while keeping it correct. Probes run whatever command your AGENTS.md/README.md "run" line declares - the platform re-reads it after every completed task - so you may rewrite in any language at any rung and the checks follow. One caveat - the server re-runs your claimed checks against the committed solution to verify the score, and its sandbox provides sh, node and python3, so behind any other interpreter your solution goes unverified rather than confirmed (it is never penalised for that). The size is the combined byte count of every file in the project except .git/, .ololo/, AGENTS.md, README.md and .gitignore, plus the declared run command itself - helper files and a leftover answer.sh count too, and moving code into the run line, or into a doc file the run line invokes, does not make it free (both are counted). Correctness is re-probed on every check with a freshly sampled string and count.

Shrink the solution to at most 62 bytes while keeping it correct. Probes run whatever command your AGENTS.md/README.md "run" line declares - the platform re-reads it after every completed task - so you may rewrite in any language at any rung and the checks follow. One caveat - the server re-runs your claimed checks against the committed solution to verify the score, and its sandbox provides sh, node and python3, so behind any other interpreter your solution goes unverified rather than confirmed (it is never penalised for that). The size is the combined byte count of every file in the project except .git/, .ololo/, AGENTS.md, README.md and .gitignore, plus the declared run command itself - helper files and a leftover answer.sh count too, and moving code into the run line, or into a doc file the run line invokes, does not make it free (both are counted). Correctness is re-probed on every check with a freshly sampled string and count.

Shrink the solution to at most 55 bytes while keeping it correct. Probes run whatever command your AGENTS.md/README.md "run" line declares - the platform re-reads it after every completed task - so you may rewrite in any language at any rung and the checks follow. One caveat - the server re-runs your claimed checks against the committed solution to verify the score, and its sandbox provides sh, node and python3, so behind any other interpreter your solution goes unverified rather than confirmed (it is never penalised for that). The size is the combined byte count of every file in the project except .git/, .ololo/, AGENTS.md, README.md and .gitignore, plus the declared run command itself - helper files and a leftover answer.sh count too, and moving code into the run line, or into a doc file the run line invokes, does not make it free (both are counted). Correctness is re-probed on every check with a freshly sampled string and count.

Shrink the solution to at most 48 bytes while keeping it correct. Probes run whatever command your AGENTS.md/README.md "run" line declares - the platform re-reads it after every completed task - so you may rewrite in any language at any rung and the checks follow. One caveat - the server re-runs your claimed checks against the committed solution to verify the score, and its sandbox provides sh, node and python3, so behind any other interpreter your solution goes unverified rather than confirmed (it is never penalised for that). The size is the combined byte count of every file in the project except .git/, .ololo/, AGENTS.md, README.md and .gitignore, plus the declared run command itself - helper files and a leftover answer.sh count too, and moving code into the run line, or into a doc file the run line invokes, does not make it free (both are counted). Correctness is re-probed on every check with a freshly sampled string and count.

Shrink the solution to at most 40 bytes while keeping it correct. Probes run whatever command your AGENTS.md/README.md "run" line declares - the platform re-reads it after every completed task - so you may rewrite in any language at any rung and the checks follow. One caveat - the server re-runs your claimed checks against the committed solution to verify the score, and its sandbox provides sh, node and python3, so behind any other interpreter your solution goes unverified rather than confirmed (it is never penalised for that). The size is the combined byte count of every file in the project except .git/, .ololo/, AGENTS.md, README.md and .gitignore, plus the declared run command itself - helper files and a leftover answer.sh count too, and moving code into the run line, or into a doc file the run line invokes, does not make it free (both are counted). Correctness is re-probed on every check with a freshly sampled string and count.